MTI Report 00-01
Norman Y. Mineta
International Institute for
Surface Transportation Policy Studies
Created by Congress in 1991
Why Campaigns for Local Transportation
Initiatives Succeed or Fail: An Analysis of Four
Communities and National Data
Mineta Transportation Institute
San José State University
San Jose, CA 95192-0219
Mineta Transportation Institute Report 00-1
Why Campaigns for Local Transportation
Funding Initiatives Succeed or Fail:
An Analysis of Four Communities and National Data
Peter J. Haas, Ph.D. (Principal Investigator)
Kristen Sullivan Massey, M.P.P.
Linda O. Valenty, Ph.D.
Richard Werbel, Ph.D.
June 2000
a publication of the
Norman Y. Mineta
International Institute for
Surface Transportation Policy Studies
Created by Congress in 1991
Technical Report Documentation Page
1. Report No.
2. Government Accession No.
3. Recipient’s Catalog No.
FHWA/CA/OR1999/19
5. Report Date
4. Title and Subtitle
Why Campaigns for Local Transportation Funding Initiatives Succeed
or Fail: An Analysis of Four Communities and National Data
7. Authors
Peter Haas, Ph.D., Principal Investigator; Kristin Sullivan Massey, M.P.P.;
Linda O. Valenty, Ph.D.; and Richard Werbel, Ph.D.-
9. Performing Organization Name and Address
6. Performing Organization Code
8. Performing Organization Report No.
00-1
10. Work Unit No.
Norman Y. Mineta International Institute for
Surface Transportation Policy Studies
College of Business, San José State University
San Jose, CA 95129-0219
12. Sponsoring Agency Name and Address
California Department of Transportation
Sacramento, CA 95819
June 2000
11. Contract or Grant No.
65W136
13. Type of Report and Period Covered
U.S. Department of Transportation
Final Report
Research and Special Programs Administration
14. Sponsoring Agency Code
400 7th Street, SW
Washington, DC 20590-0001
15. Supplementary Notes
This research project was financially sponsored by the U.S. Department of Transportation’s Research and
Special Programs Administration and by the California Department of Transportation (Caltrans).
16. Abstract
As funding from state and national sources has dwindled and demands for relief from traffic and congestion have grown, local governments and
transportation agencies are increasingly left to develop their own sources of enhanced revenues. Frequently the bid to increase available revenues
comprises a local ballot measure, enabling the citizens served by these governments and agencies to express their preferences for or against increased
taxation in support of an improved transportation system. What determines the success of campaigns in support of such ballot measures? To answer
this question, this report includes the use of two different approaches and data sources.
1) A statistical analysis of community-level characteristics. Data from localities across the nation, as well those within the state of California, that
have conducted elections for transportation tax increase are analyzed to determine what factors seem to affect the outcome of such elections.
2) Case studies of four communities that recently conducted elections for transportation tax increases (Santa Clara and Sonoma Counties in California, and the Denver and Seattle metropolitan areas). The case studies allow for in-depth, qualitative understanding of what election strategies and
other campaign elements comprise successful or unsuccessful efforts to raise local revenues.
Among the most significant findings from the statistical analysis of local elections were the following:
Efforts to fund transportation with taxes where the proportion of elderly is greater than 9 percent are more likely to succeed In communities where
the percentage of elderly is greater than 9 percent, the analysis indicates that voters may be more willing to accept local transportation taxes. However,
in communities where the percentage of elderly is less than 9 percent, transportation measures may require significantly more determined marketing
to enhance the probability of passage.
Efforts to increase sales taxes for transportation programs will be less successful in communities with higher sales taxes.
A relatively strong and negative relationship between sales tax and support for transportation tax initiatives was identified in the national election
data. This suggests that communities with relatively higher sales taxes will be hard pressed to convince citizens to support additional increases.
17. Key Words
coalitions, legislation, policy, policy
analysis, public policy, team building, transportation policy
19. Security Classif. (of this report)
Unclassified
Form DOT F 1700.7 (8-72)
18. Distribution Statement
No restrictions. This document is available to the public
through the National Technical Information Service,
Springfield, VA 22161
20. Security Classif. (of this page)
Unclassified
21. No. of Pages
140
22. Price
$15.00
Copyright © 2000 by IISTPS
All rights reserved
Library of Congress No. 99-076280
To order this publication, please contact the following:
Mineta Transportation Institute
College of Business BT-550
San José State University
San Jose, CA 95192-0219
Tel (408) 924-7560
Fax (408) 924-7565
e-mail: [email protected]
http://transweb.sjsu.edu
ACKNOWLEDGEMENTS
The Mineta Transportation Institute has received funding through the California Department
of Transportation (Caltrans) and the Research and Special Programs Administration (RSPA)
of the U.S. Department of Transportation to conduct policy related activities in the areas of
research, education, and information sharing with a focus on United States surface transportation.
In addition to the research team and interviewees, the authors thank the many individuals who
helped to complete this project, including Institute Research Director Trixie Johnson and
Communications Manager Jeanne Dittman. A special thank you goes to Brian Taylor, Ph.D.
and Assistant Professor of Urban Planning at the UCLA School of Public Policy and Social
Research, for his invaluable contribution.
Table of Contents
i
TABLE OF CONTENTS
EXECUTIVE SUMMARY
1
INTRODUCTION: THE DEVOLUTION OF TRANSPORTATION
FUNDING
Organization of this Report
The Evolution of Transportation Funding in Metropolitan Areas
3
4
5
THE IMPACT OF COMMUNITY CHARACTERISTICS ON
TRANSPORTATION FUNDING OUTCOMES
Predictor Categories
Methodology
Results
Analysis of Datasets
Directions for Future Research
21
23
25
34
41
43
CASE STUDIES: CAMPAIGNS FOR TRANSPORTATION TAX
MEASURES IN FOUR COMMUNITIES
Introduction
Prior Research on Voter Influences Pertaining to Local Transportation
Funding Initiatives
Recommendations for Obtaining Voter Approval of Transportation
Funding Initiatives
Evaluation of Recommendations for Obtaining Voter Approval of
Transportation Funding Initiatives
Case Study I
Case Study II
Case Study III
Case Study IV
CONCLUDING OBSERVATIONS AND RECOMMENDATIONS
Introduction
Critical Success and Failure Factors: Transportation and Funding
Package
ABBREVIATIONS AND ACRONYMS
BIBLIOGRAPHY
ABOUT THE AUTHORS
PRE-PUBLICATION PEER REVIEW
47
47
47
49
51
53
65
81
93
109
109
111
133
134
139
140
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
ii
Lists of Figures and Tables
LIST OF FIGURES
Figure 2-1. Type of Funding Mechanism on Transportation
Tax Measures in U.S.
Figure 2-2. Percentage Voting for Transportation as a Function of
Funding Mechanism
Figure 2-3. Percentage Voting for Transportation as a Function of
Population Growth
Figure 2-4. Percentage Voting for Transportation as a Function of
Driving Alone
44
44
45
46
LIST OF TABLES
Table 2-1. Dependent Variable
Table 2-2. Independent Variables
Table 2-3. Regression Specification for National Dataset
Table 2-4. Hypotheses for National Dataset
Table 2-5. Dependent Variable
Table 2-6. Independent Variables
Table 2-7. Regression Specification for California Dataset
Table 2-8. Hypotheses for California Dataset
Table 2-9. Average County Population for Measures that
Passed/Failed
Table 2-10. Model: Dependent Variable = Proportion Voting for
Transportation Measure
Table 2-11. Summary of results: California Data
Table 2-12. Average County Population for Measures that
Passed/Failed
Table 2-13. Summary of Results: National Data
Table 3-1. RTA Proposal
Table 3-2. PR Funding
Table 3-3. The percentage of ‘yes’ votes for each county in both the
997 GTR election and the 1995 TABOR override.
Table 4-1. Comparison of Cases on Relevant Characteristics
Pertaining to Coalition Building, a Transportation and Funding
Package, and Marketing and Communications Programs
Targeted at Voters
26
27
28
28
30
30
32
33
35
36
37
39
40
89
91
106
109
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Executive Summary
1
EXECUTIVE SUMMARY
As funding from state and national sources has dwindled and demands for
relief from traffic and congestion have grown, local governments and
transportation agencies are increasingly left to develop their own sources of
enhanced revenues. Frequently the bid to increase available revenues
comprises a local ballot measure, enabling the citizens served by these
governments and agencies to express their preferences for or against increased
taxation in support of an improved transportation system. What determines the
success of campaigns in support of such ballot measures? To answer this
question, this report includes the use of two different approaches and data
sources.
1) A statistical analysis of community-level characteristics. Data from
localities across the nation, as well those within the state of California, that
have conducted elections for transportation tax increase are analyzed to
determine what factors seem to affect the outcome of such elections.
2) Case studies of four communities that recently conducted elections for
transportation tax increases (Santa Clara and Sonoma Counties in California,
and the Denver and Seattle metropolitan areas). The case studies allow for indepth, qualitative understanding of what election strategies and other campaign
elements comprise successful or unsuccessful efforts to raise local revenues.
Among the most significant findings from the statistical analysis of local
elections were the following:
Efforts to fund transportation with taxes where the proportion of elderly is
greater than 9 percent are more likely to succeed
In communities where the percentage of elderly is greater than 9 percent, the
analysis indicates that voters may be more willing to accept local
transportation taxes. However, in communities where the percentage of elderly
is less than 9 percent, transportation measures may require significantly more
determined marketing to enhance the probability of passage.
Efforts to increase sales taxes for transportation programs will be less
successful in communities with higher sales taxes.
A relatively strong and negative relationship between sales tax and support for
transportation tax initiatives was identified in the national election data. This
suggests that communities with relatively higher sales taxes will be hard
pressed to convince citizens to support additional increases.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
2
Executive Summary
The case studies were used to identify strategic measures that advocates of
transportation tax packages could use to bolster their odds for success. Among
the key findings were the following:
• A budget cap should be identified to determine the magnitude of the
transportation package. This budget cap will help create a transportation
package that reflects community priorities for the package.
• Priorities should be based upon information from a variety of sources.
Information should be collected directly from the public through research
techniques such as focus groups and surveys. Surveys can be useful in
identifying voters’ priorities. Involvement of a citizen advisory group in the
decision-making process also can be useful in establishing priorities.
Representatives from key interest groups, such as the business community and
environmental representatives, should be included in these citizen advisory
groups.
• A combination of highway and transit improvements should be included in a
transportation package whenever it is possible to do so.
• Although opposition may not exist when effective coalition building
strategies are used, such opposition usually will arise and it may come from
surprising sources. Since the campaigns usually are of short duration, rapid
responses to opposition are needed. Contingency planning is needed to provide
effective rapid responses.
A general conclusion gleaned from the case studies is that passage of
transportation tax measures is generally an uphill battle. This is so, even in the
context of generally favorable conditions such as widespread recognition of
the need for a public response to transportation problems.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
3
INTRODUCTION: THE DEVOLUTION OF
TRANSPORTATION FUNDING
Local governments and transportation agencies are increasingly asked to
develop their own sources of funding for transit and other transportation
improvements. As funding from state and federal sources has dwindled and
local demands for relief from traffic and congestion have grown, local
governments and transportation agencies are under pressure to develop new
and creative ways to finance transportation.
Frequently, local efforts to increase transportation revenues take the form of a
local ballot measure, enabling the citizens served by these governments and
agencies to express their preferences for or against increased taxation to pay
for an improved transportation system. No matter how dire the perceived need
for more and better public transportation may be in individual communities,
passing a local tax increase is never a foregone conclusion. For example, of 57
such initiatives for public transit over the past 10 years identified for this study,
only 56 percent were successful. Similarly, 50 percent of all initiatives to fund
public transportation (transit and other improvements) in the State of
California since 1980 were approved by voters. While those percentages may
seem attractive, vast amounts of time, effort, and other resources were
expended in the many losing efforts.
What determines the success of campaigns in support of such ballot measures?
Two general explanations are typically offered. First, it is thought that the
outcome of local ballot measures is determined largely by the demographic,
socioeconomic, and geographic characteristics of communities. For example,
the likelihood of passing a transportation initiative might increase with the
population and income of a community, all other things being equal. In such
cases, the characteristics of a community could determine the success of local
campaigns for transportation ballot measures, meaning that the specific efforts
of local transportation funding measure supporters might be less important
than the environment in which they occur.
The second explanation for the relative success or failure of such measures lies
in the campaign efforts of supporters and opponents. Here the strategies,
tactics, and approaches are central to the outcomes of transportation finance
campaigns. From this perspective, the entire sequence and range of activities
associated with devising and campaigning for a transportation ballot measure
are potentially important. For example, the key to passing such a measure
might lie in the crafting of ballot language and the content of the measures
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
4
Introduction: The Devolution of Transportation Funding
themselves. Or, campaign tactics like marketing techniques and use of mass
media may be more important.
Of course, both explanations probably contribute to the success and failure of
transportation ballots. From a research perspective, however, the relative
importance of the two explanations is not known. Many forms of communitylevel demographic, socioeconomic, and geographic data can be assembled or
derived from existing data sources. Statistical techniques such as multiple
regression may be used to simultaneously measure the influence of multiple
factors in explaining the outcome of transportation tax elections. However, this
approach would probably be less likely to identify the potential effect of
specific campaign techniques and strategies.
A more appropriate way to identify and understand the impact of specific
efforts by supporters (and opponents) of a transportation measure is the case
study approach. By interviewing key officials and other individuals involved
with the campaign for or against a transportation measure – as well as using
other techniques – much can be learned about what “works” in the context of
campaigns for such measures.
The purpose of this report is to use both approaches, statistical analysis of
community-level characteristics and more qualitatively focused case studies,
to explore what determines the success of local transportation-related tax
measures. The report contains both a statistical analysis of local election data
from California (1980-1998) and the nation (1990-1998), as well as a series of
four in-depth case studies of funding initiative campaigns. The findings
provide insight into which factors seem to be the most influential in
determining the outcome of such campaigns.
Because many communities seek guidance in crafting successful transportation
tax measures, this research was conducted primarily from the standpoint of
transportation tax proponents. Nevertheless, the findings may be of use to both
proponents and opponents of such measures. The authors do not endorse any
particular proposal and the research contained in this report is intended to
provide useful information and not to advocate any particular type of measure.
ORGANIZATION OF THIS REPORT
This report is divided into four chapters. The first chapter contains this
introduction and an overview of the literature concerning the history of
funding for local transportation needs. The literature review helps to establish
the importance of localities raising their own funding for enhanced
transportation. The second chapter contains the quantitative analysis described
earlier. Both California and national data are analyzed with respect to the
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
5
impact of community-level characteristics on the outcome of local ballot
measures. The third chapter consists of the findings from case studies of local
transportation-related tax initiatives in four communities (Sonoma and Santa
Clara Counties in California, the Seattle-Tacoma region, and Denver). The last
chapter provides a collation and synthesis of the findings from both the
quantitative analysis and the case study research.
THE EVOLUTION OF TRANSPORTATION FUNDING IN
METROPOLITAN AREAS
This section reviews previous research on the devolution of decision making
and financial authority in transportation from the federal government and the
states to regional governments and counties over the past two decades, with an
emphasis on the increasing role of local finance instruments for transportation.
The political and institutional context for the devolution of transportation
decision making
This section explores the public finance of transportation facilities during the
twentieth century, with particular emphasis on one state—California—to
provide state-specific insights on devolution. Most of the conditions and
developments associated with California in this discussion apply to some
extent to major metropolitan areas in states across the country. The focus here
is on the factors affecting decision making authority and financial
responsibility for the local transportation finance plebiscites.
Financing metropolitan highways and public transit
As early as the 1920s, when the automobile emerged as more than just a
pleasure vehicle, transportation planners have struggled to cope with growing
automobile use and chronic traffic congestion in cities (Brown et al. 1998). In
the 15 years following the end of World War II, the states and the federal
government combined in an enormous financial commitment to inter-urban
and intra-urban highway development. This highway development effort, of
which the Interstate and Defense Highway System was the largest single
component, combined with new home mortgage programs and rising real
incomes to rapidly accelerate a process of residential and commercial
suburbanization which dated from the turn of the century. Rapid metropolitan
freeway development in the post-war era generally accommodated burgeoning
automobile use until funding began to run short in the late 1960s. Significant
growth in highway development costs and increasing environmental permit
requirements on one hand and a dependence on inflation-vulnerable fuel taxes
levied on a per-gallon basis on the other combined to vise metropolitan
freeways in a cost-revenue squeeze from which they have yet to emerge
(Taylor 1995).
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
6
Introduction: The Devolution of Transportation Funding
While metropolitan road and freeway development has dramatically declined
since the mid-1960s, automobile travel has continued to grow. This growth is
due both to increases in population and to increased levels of automobile use.
The predictable result, of course, has been chronic traffic congestion in cities.
In 1996, 22 percent of lane miles in California’s corridors of high economic
significance were rated Level of Service (LOS) F, a breakdown of traffic flow
to stop-and-go conditions (Commission on Transportation Investment 1996).
Throughout much of the 1980s and 1990s, California voters and local
authorities have identified transportation as one of their most pressing
concerns. As a result of these concerns, voters have demanded greater
participation in the transportation decision-making process and local officials
have sought greater control over transportation planning outcomes.
Revenues for highway construction have not kept up with demand
The number of vehicle miles traveled (VMT) has increased steadily since the
1910s. Particularly since 1980, demand for transportation has increased
considerably in California. Although VMT have increased more than 50
percent since 1980, delay due to congestion has more than doubled (ACIR
1988). Despite the increase in demand for highway access, the amount spent
for California highway right of way has decreased substantially from a high of
$.90 per 100 VMT in 1956 to only $.10 per 100 VMT in 1988 (Taylor 1995).
The inability of federal and state governments to spend more on transportation
infrastructure is due largely to the inability of the main funding mechanism, the
gas tax, to keep up with inflation (Brown et al. 1998). The gas tax, currently set
at just over 36 cents per gallon, has a built-in “sunset clause” associated with
inflation and increasing fuel efficiency. That is, without yearly increases, the
purchasing power of the monies derived from the gas tax decrease as inflation
and vehicular fuel efficiency increase.
Additionally, transportation infrastructure costs have risen. This rise is due to
the significant increase in the cost of rights-of-way in urban areas and larger
numbers of federal and state mandates for transportation planning, such as the
requirement for environmental impact review. These costs have occurred
without corresponding increases in funding to pay for them and have greatly
expanded costs for providing transportation infrastructure. In sum, the inability
of federal and state governments to exact yearly gas tax increases, when
combined with continuing increases in the construction costs for transportation
infrastructure, has resulted in a revenue shortfall (Taylor 1995). It has been
increasingly difficult to finance needed transportation improvements with
revenue from these funding mechanisms. As a result, voter dissatisfaction with
state and federal governments’ ability to find solutions to transportation
problems has also increased (Brown et al. 1999).
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
7
Cutbacks in federal aid compel local action
During the 1980s, the federal government’s emphasis on greater local and
regional responsibility for transportation decision making was limited
considerably by significant cutbacks in financial aid (Zoller & Capizzano
1997). As a result, local jurisdictions were encouraged to seek local financing
to fund transportation projects. Clearly, the perception that local transportation
finance ballot measures are more politically feasible than statewide measures
has contributed to rising levels of funding available to localities (Brown et al.
1998). Like cities and counties in most states, the cities and counties in
California are not allowed to impose a gas tax for transportation (ACIR 1988).
However, several counties, desperate to avoid state spending limitations on
transportation, persuaded legislators to pass SB878 in 1986. SB878 authorized
nine San Francisco Bay Area counties plus Fresno to place dedicated sales
taxes of up to 1 percent on the ballot to finance transportation projects to be
approved by a simple majority (Colman 1987). In several California counties,
the public has supported the notion of local control and generally believed in
the increased availability of local tax revenue, and so were able to successfully
pass local sales taxes dedicated entirely to transportation (Brown et al. 1998).
To better contextualize the devolution of transportation funding in
metropolitan areas, the sections below offer a short history of transportation
finance in one state, California, as an example of the shifting institutional roles
and responsibilities for metropolitan transportation planning and finance.
Transportation finance in California
1. 1910-1930: Increasing demand for road construction; State sponsored
financing
Emergence of the automobile as more than just a pleasure vehicle
In the early years of the highway program, the automobile was regarded as a
pleasure and recreational vehicle rather than an important means of
transportation. Consequently, highways consisted of relatively short sections,
built from city to countryside (Weiner 1986). With the surge in automobile
ownership between 1900 and 1919, automobile owners organized and lobbied
for better roads. Doubting the capacity of local government to generate
sufficient revenue or to have adequate construction expertise, automobile clubs
lobbied state legislature for direct state involvement in road financing and
construction. In California, that political pressure paid off when the legislature
enacted the State Highway Act of 1909 that created a 3,082 mile state highway
system.
The system in California, as in many other states, relied on bond financing to
be retired with general tax revenues. By 1922, however, nearly 44 percent of
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
8
Introduction: The Devolution of Transportation Funding
all California state revenue had been devoted to the state highway program.
New bond issues were not able to keep up with the demand for new roads. As a
result, innovations in the current funding mechanism were investigated (Brown
et.al. 1999). In 1923, a gas tax was adopted in California as it was in that era by
many other states. Due to its popularity, the gas tax became the primary
instrument by which revenues were raised for the construction and
maintenance of the state’s highway system until the 1990s.
These funding methods, however, were not without controversy. By the mid1920s, resistance to the gas tax occurred in several states around the country as
these states began to use the gas tax money for non-road-building purposes. As
a result, the petroleum industry and automobile clubs waged successful
campaigns to ensure that state tax revenues were not diverted from highway
purposes (Brown 1998).
Recognition for the need to plan transportation projects
The need for a systematic approach to highway planning was recognized in the
1930s as rapid growth in automobile ownership and highway travel placed
increased demands on an inadequate highway system (Weiner 1986). In 1934,
with the passage of the Federal Aid Highway Act, Congress created a
cooperative arrangement between the U.S. Bureau of Public Roads (now the
Federal Highway Administration) and the state highway departments. Demand
for road construction continued but the arrival of the Depression put many
Californians out of work and imperiled state and local government revenuegenerating sources. The legislature responded with increased financial aid in
urban areas designed to put more Californians to work.
2. 1930-1960: Increasing demand for road construction; Corresponding
financing available
Great public, business and political support for federal investment in highway
construction
The public and many businesses were strong supporters of federal investment
in highway construction. In 1947, the California Report on Highways, Streets
and Roads, reported that “the present system does not have sufficient capacity
to move existing volumes of traffic economically, safely, or conveniently”
(p.4). California’s response to the need described in the report was to ensure
that key segments of the system were rebuilt “to modern standards of lane
capacity, alignment and grade; and all tested devices for minimizing traffic
hazards... be built into the system without delay.” (p.4)
Nearly everyone favored the construction of the interstate highway system and
the dramatic increase in federal aid that would support it (Brown et al. 1998).
At that time, the advantages of automobility were so great and the potential
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
9
problems (community disruption, air pollution, noise pollution, ground water
pollution, and vulnerability to international oil shocks) so speculative, that
autos and highway development were almost universally popular. Highways
were widely associated with economic benefits of congestion relief and the
pleasurable experiences of free-flowing driving and easy access to the
countryside. Furthermore, individual Americans had substantial investment in
their cars and were “amenable to the suggestion that good roads were essential
for their full utilization.” (Altshuler 1978) Public support for highways was in
part due to the lack of competition. Privately-owned public transit companies
were widely unpopular in many cities, and offered little service in small towns
or rural areas (Jones 1985). The automobile, in contrast, could reach rural areas
and with the construction of additional roads would provide even greater
freedom from dependence of poorly maintained private streetcar systems.
In addition to general public support, there was great industry support for
highway construction. The automobile, trucking, oil, and rubber industries, and
the labor unions campaigned and lobbied for new roads, arguing that not only
would highway construction serve mobility and defense objectives, but would
also help to attain many of the employment and economic objectives of the
country. Such claims were not, as it turns out, farfetched; roughly one-sixth of
all American businesses employing one-seventh of American employees are
involved in the highway and automobile industries (Altshuler 1978). Highways
and highway expansion had support from a significant proportion of the
nation's largest companies – all of the top seven and 12 of the top 15
companies in Fortune’s annual ranking of the largest industrial companies in
America had ties to automobile, oil, and steel companies (Altshuler 1978).
Furthermore, there was great political support for highway construction.
Businesses and workers that would benefit were spread fairly evenly across
congressional districts compared to transit options which were concentrated in
predominantly urban areas. About 60 percent of all transit patronage in the
early 1990s were on the 10 largest public transit systems. (Taylor and
McCullough 1998).
Growing demand that road money be used exclusively for roads
There was growing debate between the 1930s and 1950s about government’s
intent in financing transportation infrastructure using the gas tax. In California,
as in all states, decisions about local transportation projects had been
controlled at the state and federal level for most of the century. In California,
the California Department of Transportation (Caltrans) was the agency
primarily responsible for controlling highway construction projects. In fact, the
California Constitution generally required that the state commit transportation
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
10
Introduction: The Devolution of Transportation Funding
revenues to the construction and maintenance of highways, roads, and streets,
or in certain cases, to the construction of urban rail transit projects.
Near the close of World War II, a system of highways designated as the
“National System of Interstate and Defense Highways” was authorized in the
Federal Aid Highway Act of 1944. The final route choices recommended by
the U.S. Bureau of Public Roads were influenced as much by strategic
necessity and other issues such as population density, concentrations of
manufacturing activity, and agricultural production, as by existing and future
traffic density. The importance of the system within cities was recognized;
however, it was not the intent of the program that these highways serve urban
commuter travel demands in major cities. In fact, it was determined “both
locally and nationally, to recognize the recommended systems as that system
and those routes which best and most directly join region to region and major
city to major city” (Weiner 1986). In 1947, a 37,700 mile network of the most
heavily traveled routes was adopted for inclusion in a National System of
Interstate Highways. Funds appropriated, however, were at very low levels
(with a 50 percent federal share) through 1956 (Weiner 1986).
When World War II ended, California faced a very serious shortfall of
highway revenues as materials and manpower were diverted to the war effort.
However, highway traffic continued to grow, especially heavy vehicle traffic
used to transport war-related materials which caused enormous damage to
highways and bridges designed to carry much smaller loads. In the face of the
enormous postwar maintenance and construction needs, the legislature
established a series of special committees to analyze the situation. In
California, the Collier-Burns Act of 1947 was passed, increasing gas taxes to
4.5 cents per gallon. The act also created a Highway Users Tax Fund in
California that ensured that taxes collected would be used solely for highway
and street construction.
While highway departments were placing a major emphasis on arterial routes,
city street congestion was steadily worsening. It was in this atmosphere that the
Committee on Urban Transportation was created by Congress in 1954. Its
purpose was “to help cities do a better job of transportation planning through
systematic collection of facts...to afford the public the best possible
transportation at the least possible cost and aid in accomplishing desirable
goals of urban renewal and sound urban growth.” They developed a guidebook
in 1958 to help local officials establish an orderly program of urban
transportation planning. It stressed the need for “cooperative action and the
development of transportation systems in keeping with the broad objectives of
community development.” (Weiner 1986)
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
11
In 1957, the Federal Aid Highway Act increased the authorized network to
41,000 miles intending to link 90 percent of the cities with populations of
50,000 or more. Specifically, it authorized $24.8 billion over 13 years from
1957 to 1969, and this time at a much higher federal match – 90 percent. The
companion act, the Highway Revenue Act of 1956, increased federal taxes on
gasoline and other motor fuels, and increased excise taxes to create a Highway
Trust Fund that would receive tax revenue dedicated solely for highway
purposes (Weiner 1986). Thus, legislation authorized in the late 1950s, marked
the beginning of the modern era of highway finance with its temptingly large
federal matching grants.
While the Collier-Burn Act jump-started freeway development in California,
in most other states development was simply too costly for state or local
jurisdictions to fund prior to the passage of the Highway Revenue Act of 1956.
As a result, and for almost two decades after 1956, the focus of attention in the
urban transportation policy arena was on major capital projects (Altshuler
1979).
By 1959, the deluge of new federal highway money, combined with
California’s own surpluses ($110 million in 1957 and $145 million in 1958)
prompted the California state legislature to embrace a superhighway system of
its own, the Division of Highways’ 12,000 mile Freeway and Expressway
System plan. The state’s program “enjoyed nearly universal support because it
provided 12,241 miles of high quality roads to California motorists without
asking them to pay a penny more in taxes than they were already paying”
(Brown et al. 1998). From the 1930s through the 1950s, growing awareness of
transportation-related urban problems led to an increase in funding for
metropolitan highways
3. 1960-Present: Increasing demand for transportation; cost inflation,
revenue erosion
Local control over highways has gradually increased, through a series of state
and federal policy decisions, over the past two decades. Beginning in the
1980s, the use of county sales tax supplements was an early step in the
direction of local control.
Recent recognition of unmet transportation needs
By the late 1960s, the urban transportation planning process was criticized on a
number of fronts. Criticism was directed at the fact that the planning process
did not adequately treat social and environmental impacts, was not multimodal, and focused almost exclusively on long-range time horizons. Further,
technical procedures were too cumbersome, time-consuming, and rigid to
adapt to new issues quickly (Weiner 1986). As a result, in 1962, the Federal
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
12
Introduction: The Devolution of Transportation Funding
Aid Highway Act declared that it is in “the national interest to encourage and
promote the development of transportation systems embracing various modes
of transport in a manner that will serve the states and local communities
efficiently and effectively.” (Weiner 1986)
To address concerns that the federal government was not meeting its
responsibilities in providing basic mobility, the Urban Mass Transportation
Act was enacted in 1964. It was designed to ensure that they provided funding
to maintain service quality through transit expansion and improvements; but
also mandated that “fiscal discipline” be guaranteed by requiring that operating
costs be covered by fare box revenues (Altshuler 1978). For example, in 1962
voters in the San Francisco Bay area were presented with the BART proposal
which assured that no operating subsidies would be required (Altshuler 1978).
In 1970, the Urban Mass Transportation Assistance Act provided the first longterm commitment of federal funds for mass transportation, a federal
commitment of at least $10 billion over a 12-year period (Weiner 1986).
At the 1974 Mt. Pocono Conference on Urban Transportation Planning,
decision makers recommended close coordination of planning efforts as a
means of achieving orderly development of urban areas and relating the
planning process more closely to decision making processes at all levels of
government. It urged that urban planning be strengthened through state
enabling legislation and bolstered by equitable local representation. The
conference report went on to urge that “this more inclusive kind of planning be
supported by flexible funding from the federal government.” This was to be
done to prevent preference for any particular mode and to avoid an imbalance
in specific urban transportation decisions that could be contrary to local goals
and priorities (Weiner 1986).
Additional program requirements strained budgets
In California, the California Environmental Quality Act (enacted in 1970)
altered the context in which highway design was to be conducted. In general it
required that before a construction project proceeded, its effect on the natural
environment must be assessed and where feasible, effort should be made to
mitigate any negative effects. Similarly, any negative effects on the social or
economic fabric of a community must be identified and attempts made to
mitigate when feasible. Finally, Environmental Impact Review (EIR)
statements were required to be circulated among interested agencies and
private groups. This has resulted in a large increase in the number of
participants in highway investment decisions, not all of whom share the
common belief in the benefits of highway projects. The process is both
complex and time consuming, perhaps made more so by the general perception
that the interpretation and accommodation of the requirements of this body of
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
13
law was difficult (Commission on State Government 1983). Finally, the
Federal requirements for accessibility by the disabled and elderly required
manpower and time, and resulted in a need for additional financing (Altshuler
1979).
New sources of financing available to localities
In 1997, the California legislature passed Senate Bill 45, which granted
regional control over 75 percent of the funds available for the State
Transportation Improvement Program (STIP). In addition, counties began to
use sales taxes to finance transportation (Brown, et al., 1998). In the 1980s,
counties were given the option to raise additional sales tax revenue for
transportation purposes resulting in 18 counties (as of 1999) which had enacted
local sales taxes dedicated to transportation (Brown et al. 1998). By 1996-97,
they had raised almost a quarter as much revenue as the state fuel taxes had
raised for highways (approximately $400 million).
Property taxes have long been another important source of transportation
funding, particularly for local streets and roads. In 1995, for example, they
contributed $145 million toward local streets and roads (Brown et al. 1998).
In 1981, SB 215 instituted county minimum expenditure requirements which
mandated that at least 70 percent of the funds in each county be distributed
among the counties on the following basis: 75 percent of this percentage is
distributed according to each county’s population relative to the total
population in its county group; the other 25 percent is distributed according to
how many state highway miles each county has open to travel (Commission on
California State Government 1983).
Revenues for highway construction decline
From the mid-1960s through the late 1980s, highway construction and
maintenance costs increased, while available revenue declined in real dollar
terms. During the 1960s, cost inflation began to occur, but legislators were able
to mitigate this problem by extending the build-out phase of various projects
rather than by enacting substantial tax increases or reducing the scale to
program commitments. By the end of the 1960s, the ability of the gas tax to
respond automatically to inflation became a serious issue. Mild inflation in the
late 1960s due to the Vietnam War and severe inflation in the 1970s due to the
oil crisis further reduced the purchasing power of gas tax revenue. Between
1964 and 1982, the gas tax entered a period of substantial decline during which
its buying power was again eroded, this time by well over two-thirds. Today, if
the purchasing power of the gas tax were lowered to (inflation-adjusted) 1964
levels, the current gas tax would have to be doubled (Taylor 1995).
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
14
Introduction: The Devolution of Transportation Funding
Additionally, through the 1960s, annual gasoline consumption (and hence fuel
tax revenue as a result of gas taxes), exceeded increases in construction costs.
This began to change in about 1970 as construction costs began to rise. And,
California’s return from federal highway tax dollars dropped from a high of 85
cents of every dollar to little more than 60 to 65 cents in the 1980s
(Commission on California State Government 1983).
In the 1970s, the gas tax was enormously affected by energy conservation
policies. The new concern over energy policy began with the oil embargo of
1973. In only a few short years, gasoline and other energy prices tripled and
concern over America’s energy dependence on foreign oil became a leading
public issue. A particularly salient legacy of the Energy Policy and
Conservation Act of 1975 with regard to transportation financing, was the
establishment of the vehicle fuel efficiency standards. As a result, automakers
began building more fuel-efficient vehicles (Brown et al. 1998). More fuelefficient cars reduced tax revenue even further by increasing the distance
commuters could travel on a gallon of gas.
Between 1964-1983, California did not change the gas tax, while at the same
time, cost inflation for transportation construction soared. Between 1952 and
1967 the annual rate of increase in highway construction was 2 percent.
Between 1968-1973, it was 10-12 percent; between 1975-1980 it was 18
percent (Commission on California State Government 1983).
In California, the 1960s saw general inflation at an average of 2.4 percent vs.
8.2 percent for highway construction. In the 1970s, general inflation averaged
8.7 percent vs. 12.1 percent for highway construction. The rapid construction
cost inflation was associated with a general rise in maintenance and
construction costs, an upscaling of highway design standards, increased urban
right-of-way costs, and the ongoing costs of complying with federal
environmental legislation enacted in the 1970s (Taylor 1995). As a result, the
10-year approach to highway development was curtailed. In 1977, the
legislature initiated new reforms when it enacted the Alquist-Ingalls Act. It
created the State Transportation Improvement Program (STIP), a process
which established the requirement that Caltrans develop a five year schedule of
highway and other transportation investments (Commission on California
State Government 1983).
In addition, states began to rely on the sales tax to finance transportation
because of the need for additional revenue and the sales tax’s inability to keep
pace with inflation (Brown et al. 1998). In 1971, California enacted the
Transportation Development Act which extended the state sales tax to
gasoline, thus making new revenues available for urban mass transit and streets
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
15
and roads in rural areas. However, despite this increase in revenue, in general
the state and federal governments’ response to declining fuel consumption and
rising inflation was to delay transportation projects. In California, Caltrans
extended the scheduled completion of projects into the 1990s and reduced or
eliminated some highway projects entirely (Brown et al. 1998).
In 1979 the California Legislative Analyst’s Office, the California
Transportation Commission (CTC), and Caltrans began warning of an
impending financial crisis in transportation (Brown et al. 1998). In 1972, 13
years after the creation of the system, there was a backlog of $10 billion in
projects; and the backlog was projected to reach $20 billion by 1980. As a
result, Caltrans announced that the state faced a $915 million shortfall in the
five year State Transportation Improvement Program. The State legislature
responded by passing Senate Bill 215 in the 1981-1982 session. Drivers license
fees were increased to $6.75, vehicle registration fees were increased to $22,
and state, gas, and diesel taxes were increased from seven to nine cents (Brown
et al. 1998). However, there was a widespread belief that this too was only a
stopgap measure and that periodic increases would be necessary to keep pace
with inflation.
By 1990, the California Transportation Commission and Californians for
Better Transportation perceived that statewide transportation needs were close
to $40 billion (Brown et al. 1998). The October 1989 earthquake, which
devastated portions of San Francisco and the Monterey Bay areas, provided the
motivation to authorize a series of tax and bond increases to improve
transportation infrastructure with two bills. SB 300 increased the gas tax to 18
cents per gallon; AB 471 raised commercial vehicle weight fees by nearly 60
percent. These were passed along with special state sales tax increases to repair
earthquake damage and an earthquake retrofit bond measure. SB 300 and AB
471 were developed to be components of a 10-year financial package called
the Transportation Blueprint for the 21st Century. The Blueprint also called for
bond measures, which authorized the state to issue $1 billion in general
obligation bonds to pay for the construction of rail systems.
Since these tax increases exceeded constitutional limitations, legislation had to
be submitted to voters to amend the California constitution. Voters
overwhelmingly passed the amendment, named Proposition 111, the Traffic
Congestion Relief and Spending Act of 1990. Voters also passed Proposition
116, which authorized the state to sell $1.99 billion in general obligation bonds
to pay for intercity and commuter rail systems. The propositions increased
projected transportation revenues by some $18.5 billion over 10 years through
the gas tax, diesel and vehicle weight fee increases. However, this met only a
part of the actual identified need ($40 billion).
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
16
Introduction: The Devolution of Transportation Funding
A comparison of revenue generated per VMT in 1960 to revenue generated per
VMT in the late 1990s, indicated that it would be necessary to increase state
gas taxes by 29.9 cents just to match revenue generated per vehicle mile
traveled in 1960. That is, a fuel tax of nearly 48 cents per gallon would be
required to account for the erosion of the state fuel tax due to both travel
increases and inflation over the period between 1960 and 1997. As a result, the
cost to drive (per mile) is now much less than nearly four decades ago,
although today more people rely upon and expect the safety and accessibility
of roadways than in 1960 (Brown et al. 1998). Furthermore, unless taxes are
increased, altered, or replaced at either the state or federal level, the purchasing
power will continue to erode (Brown et al. 1998)
Federal government supports devolution
By the 1960s, a growing number of urban legislators criticized the urban
transportation planning process for failing to adequately address social and
environmental problems by over-emphasizing development and ignoring
deteriorating public transit systems (Weiner 1986). To address these problems,
a series of legislative initiatives were adopted to increase subsidies to public
transit and, later, to increase local decision making authority.
Until the 1960s, the federal government established separate programs for
public transit and highways. In the 1960s, the federal government began to
develop objectives that were not mode specific. In 1962, a joint report on urban
mass transportation identified common objectives for highways and mass
transit: “Major objectives of urban transportation policy are the achievement of
sound land-use patterns, the assurance of transportation facilities for all
segments of the population, the improvement of overall traffic flow, and the
meeting of total transportation needs at minimum cost. Only a balanced
transportation system can attain these goals and in many urban areas this
means an extensive mass transportation network fully integrated with the
highway and street system (Weiner 1986).”
The objective of integration with the street system required the participation of
local transportation authorities. Toward this end, the Federal Aid Highway Act
of 1962 required that the approval of any federal aid highway projects in
urbanized areas of 50,000 or more in population be based on a continuing and
comprehensive urban transportation planning process carried out cooperatively
by states and local governments (dubbed the “3C” process) (Weiner 1986).
Not until the late 1960s, however, did broad support for local participation
begin to emerge. In 1970, the Federal Aid Highway Act further strengthened
the influence of local officials in urban areas with the statement “No highway
project may be constructed in any urban area of 50,000 population or more
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
17
unless the responsible local officials of such urban area... have been consulted
and their views considered with respect to the corridor, the location, and the
design of the project (U.S. Department of Transportation in Weiner 1986).”
In the early 1980s, the Reagan administration wanted to decrease the federal
role in transportation even further (Smerk 1991). In 1991, federal legislation in
the form of the Intermodal Surface Transportation Equity Act (ISTEA) gave
regions more discretion over federal funds than ever before. Like previous
acts, it authorized funding for highway programs, bridges, tunnels, rail, and
safety. ISTEA was revolutionary, however, because it placed a new emphasis
on intermodalism, encouraged funding flexibility, and encouraged greater local
decision making. In particular, it allowed a large portion of highway funds to
be used for mass transit (Brown et al. 1998). Rather than focus on a national
highway system, ISTEA’s authors envisioned a National Intermodal
Transportation System linking intermodal facilities (Anderson 1995).
ISTEA implicitly promoted intermodal and multimodal transportation systems
by emphasizing funding flexibility across modes (Anderson 1995). It delegated
transportation planning and programming responsibilities explicitly to state
and local governments. ISTEA encouraged greater local decision making by
devolving transportation planning and programming responsibilities to state
and local governments and by giving Metropolitan Planning Organizations
more responsibility for developing long-range transportation improvement
plans for their regions (Anderson 1995).
ISTEA made changes to the planning process, the types of projects that could
be considered, and the financing requirements. In order to encourage local
concerns and a variety of perspectives about the effects of transportation,
ISTEA required that all Metropolitan Planning Organizations (MPOs) consider
the “overall, social, economic, energy, and environmental effects of
transportation decisions (Zoller & Capizanno, 1997).” Toward this end, ISTEA
required that the planning process change to include greater community
involvement (public agencies, citizens, and private providers of transportation)
in transportation decisions. About $155 billion was made available in fiscal
years 1992 to 1997.
ISTEA represented a change in how surface transportation was planned and
funded in the United States. It was revolutionary because it placed a new
emphasis on intermodalism and it afforded greater flexibility to states and
localities, allowing them to be the primary determinants of how transportation
priorities were set and how transportation funds were spent. In fiscal year
1992, a total of $16.9 billion was obligated while transfers from highways to
transit totaled $302.4 million. To instill flexibility, FHWA simplified
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
18
Introduction: The Devolution of Transportation Funding
procedures for transferring Surface Transportation Program funds between
highways and transit. ISTEA encouraged states to develop cost-sharing
partnerships with the private sector (Larson 1993) and placed a strong
emphasis on efficiency, productivity, and environmental responsibility (Miller
1992). Most recently, the passage of the Transportation Equity Act for the 21st
Century in 1998 (TEA-21) redefined intergovernmental relationships and
continued the policy of giving regions greater transportation decision making
authority vis á vis the federal government.
Voter resistance to new taxes
Since the 1970s, Americans have become increasingly skeptical about the local
benefits of growth (and related funding priorities that encourage growth, such
as transportation). As a result, voters have been resistant to new taxes for
transportation (Altshuler 1993). Furthermore, voters have become more
suspicious of the state’s ability to govern effectively. In California, voters
passed Proposition 13 in 1978 in part to limit property tax increases to pay for
what they believed was suspect public investment. Proposition 13 rolled back
property tax valuations to their 1975 valuations initially and set the property
tax rate at 1 percent. As long as property is held by the original owner, it is
valued a the 1975 rate plus a 2 percent per year inflation allowance. This trend
has continued throughout the 1990s and has been replicated in many other
states and localities. Throughout the 1990s, increasingly local control of
government has won increasing acceptance for several reasons: Republicans
have promoted the idea of government that is smaller, more efficient, and
closer to the people; the Supreme Court has increasingly ruled in favor of
state’s rights; and robust economy has reduced the perceived need for big
government (The Economist 1998).
Voters want government accountability
In the early 1900s, Californians wanted a state government that intervened in
the welfare of its citizens. However, voters became skeptical of the state
government’s ability to govern effectively as local benefits became less
apparent and negative impacts became more obvious (Altshuler 1993). In
particular, during the mid-1970’s, transportation issues began to reach a
critical stage where citizens publicly resisted the negative impacts of the
automobile—including air pollution, community disruption, urban sprawl, and
vulnerability to reliance on international oil. Furthermore, local jurisdictions
began to question the wisdom of relying on federal and state funding to finance
and build road projects, particularly when these were believed to contribute
significantly to negative impacts on the environment and land use. In such an
environment, legislators became reluctant to raise state gas taxes to fund
highway projects (Brown et al. 1998). More recently, term limits imposed by
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Introduction: The Devolution of Transportation Funding
19
voters have decreased the expertise of legislators, especially in technical areas
such as transportation. As a result of these developments, state and federal
government control in transportation decision making has steadily eroded, and
in response, local governments’ authority has increased.
In the last 10 years, two changes in the political system have further removed
transportation decision making from the hands of elected officials and put it
into the realm of voter initiatives and public polls. The first change is the
imposition of term limits. Term limits have reduced the technical expertise of
legislators simply because of natural limits on the amount of technical
transportation knowledge that legislators can develop within their term of
office. Second, elected officials are increasingly relying on public opinion to
shape and confirm their own policy choices (Brown et al. 1998). As a result,
public leaders have begun to hand decision making authority on highway
finance policy to the voters.
In California this trend has been most recently demonstrated by the fact that
the state legislature required voter approval of the last fuel tax increase. And
the population of California will be consulted continually as mandated by the
legislative requirement that voters approve of any county sales taxes and
regional gas taxes. Citizens of metropolitan areas across the nation are being
asked to do the same thing.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
20
Introduction: The Devolution of Transportation Funding
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
21
THE IMPACT OF COMMUNITY CHARACTERISTICS ON
TRANSPORTATION FUNDING OUTCOMES
For the most of the 20th century, local transit and highway transportation
authorities relied heavily on state and federal financial aid. This has been
especially true since the mid 1940s, as the demand for roads increased due to
the confluence of rising incomes, increasing suburbanization, and increasing
automobile ownership. As a result, federal highway aid, in particular, the
Interstate Highway Act of 1956, funded the largest-scale public works project
in history (Lewis & Sprague 1997). At the same time, transit service declined
in part because there was a decline in transit demand as a result of increased
competition from the automobile but, more importantly, because transit costs
had risen as a function of increased wages (Altshuler 1979). These trends led to
the Urban Mass Transportation Act of 1964, which initiated funding of public
transit capital investment that has continued to the present (Lewis and Sprague
1997). Although the federal government originally required that transit
operating costs be covered by fare box revenue, eventually the federal
government began to subsidize operating costs as well. This federal
subsidization occurred as public perception of mass transit evolved and the
public began to see transit as serving a national interest by providing mobility
to the transit-dependent (Altshuler 1979). As a result, over the past 30 years
local highway and transit transportation authorities have operated within a
fiscal environment where they rely heavily on state and federal government
subsidies to maintain service quality.
The challenge for local transportation authorities to meet demand and cover
costs has been made even more difficult since the 1980s, when President
Reagan developed objectives to decentralize financial and decision making
authority (Weiner, 1986). This trend continued through the 1990s with the
passage of the Intermodal Surface Transportation Act (ISTEA) and the
Transportation Equity Act (TEA-21) for the 21st century, which further
devolved decision making to the local level (Zoller & Capizanno, 1997).
Although increased flexibility in funding and innovative project selection is
encouraged through ISTEA and its successor, TEA-21, these tools have not
been successful in increasing the total amount of money available for either
local transit or highway transportation authorities.
Increased flexibility has increased the revenue used to finance capital transit
projects, especially for rail, but not to fund operating expenses (Federal Transit
Administration, 1998). The total transit operating subsidies have not changed
since 1988, resulting in a decline in federal operating assistance at an inflation
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
22
The Impact of Community Characteristics on Transportation Funding Outcomes
rate of 3.5 percent per year (Federal Transit Administration 1998). By 1996,
fare box revenues covered only one-third of operating expenses, with federal,
state, and local revenues covering the rest (Levine 1998). In addition, despite
the increase in demand for highway access, the amount spent for highways has
decreased substantially. In California, highway right of way decreased from a
high of $.90 per 100 VMT in 1956 to only $.10 per 100 VMT in 1988 (Taylor
1995), due largely to the inability of the gas tax to keep up with inflation
(Brown et al. 1998).
As a result of decreasing federal and state assistance for highway and transit,
local transportation authorities have increasingly had to rely on local funding
mechanisms to finance improvements and maintain service quality (Levine
1998). Specifically, local funding mechanisms available to local authorities
have generally been restricted to local taxes, a process that has typically
required voter approval. For example, since 1985, California counties have
relied primarily on local sales tax as a major source of revenue for local
transportation projects (LAO 1996). As of 1996, 17 of California’s 58 counties
had implemented at least a half cent local sales tax to fund local transportation
(LAO 1996), raising almost a quarter as much revenue as the state fuel taxes
raised for highways, approximately $400 million/year (Brown et al. 1998).
As a result, the number of local transportation tax measures that have required
voter approval has increased. In California, for example, the cumulative
number of local transportation measures that have required voter approval
increased substantially during the 1980s, from only three measures in 1980 to
over 60 measures by 1998 (see Figure 2-1). Furthermore, these measures have
affected an increasing number of California residents. Since 1980, the
percentage of Californians affected by the passage of voter approved
transportation measures has increased from 31 percent in 1980 to over 80
percent in 1998 (see Figure 2-2).
Despite the increase in local transportation tax measures, obtaining voter
approval for costly transportation projects been a challenge, in part, because
pollsters and planners have typically relied on anecdotal evidence or small
samples and limited experience to identify factors that may influence voters.
As local communities increasingly rely on local voters to approve financing for
transportation improvements, it has become more important to move beyond
anecdotal evidence to an enhanced understanding of factors which affect voter
reaction to transportation tax measures.
A few previous studies have found that several community characteristics may
be related to the successful passage of transportation and transit tax measures
(see, e.g., Levine et al. 1998). As enhanced understanding is the goal of the
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
23
present study, the validity of factors identified by previous studies is examined.
In addition, the scope of these previous studies is expanded by examining two
datasets for which voting outcome data were available: a nationwide sample of
57 city and county transit tax elections since 1990; and a comprehensive
sample of 63 California county transportation tax elections since 1980.
Based upon the predictors that were identified in previous studies (described in
detail below) seven categories of demographic variables were selected. These
variables were tested to determine their effect upon the number of citizens
voting for transportation tax measures.
PREDICTOR CATEGORIES
Age
Based upon interviews with transportation planners in California, Colman
(1987) concluded that younger voters may be more willing than older voters to
vote for transportation taxes. Harmatuck (1973) offered one explanation for a
similar finding based on surveys with voters in Madison, Wisconsin. He
suggested that because transit benefits may take years to be realized, older
voters, for whom transit taxes (and their benefits) often extend beyond their
lifetimes, may be less likely to vote for transit taxes. Others have argued that
older voters may not experience the need for transportation improvements as
urgently as younger voters because they are better able to modify their
commute schedules to avoid congestion during peak hours.
Income
In the same study, Colman (1987) concluded that higher income voters may be
less willing to vote for transportation taxes than lower income voters.
Similarly, Harmatuck (1973) found that those with lower incomes tended to
react more positively to transit tax issues. Harmatuck attributed the support of
transit taxes by lower income individuals to potentially increased use of the
service.
General economic conditions
Several planners interviewed by the journal Passenger Transport reported that
it was their perception that economic variables have affected the outcome of
transit tax measures when adverse economic circumstances have existed
(Passenger Transport 1980-1996). They suggested that poor economic
conditions, including local or national economic recessions and economic
instability, even when perceived generally, can affect willingness of voters to
support tax increases, even for needed transit.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
24
The Impact of Community Characteristics on Transportation Funding Outcomes
Taxes
Several researchers have found that sentiment regarding taxes has negatively
affected the outcome of transit tax measures. Specifically, in a poll of
California voters, Citrin (1999) found that nearly 40 percent of voters
expressed great discontent with increasing taxes. In addition, researchers found
that sales taxes are more positively evaluated by voters when they are a
component of transit tax measures. Specifically, in a historical examination of
the issue, Gomez-Ibanez (1999) concluded that use of sales taxes may be more
acceptable among voters in part because it more clearly links charges to users.
Similarly, detailed case studies of local elections in 11 states revealed that the
sales tax is more popular with voters than either the gas tax or property tax
(Beale, 1996). Researchers also found, predictably, that the larger the amount
of the tax, the less likely its passage (Middleton 1998, Beale 1996, Colman
1987). Finally, findings also indicated that the shorter the duration of the tax,
the more likely its passage (Beale 1996, Middleton 1998).
Highly automobile-oriented population
A few authors have suggested that the considerable investment by the majority
of voters in owning an automobile has discouraged voters from voting for tax
increases to fund public transit. Specifically, a highly automobile-oriented
population was found to negatively affect the passage of transportation tax in
California (Colman 1987). Similarly, Harmatuck (1973) found that a higher
degree of suburbanization negatively affected the passage of a transit tax in an
examination of voting behavior. Harmatuck attributed these results to selfinterest: those who are most likely users of the system have the greatest
propensity to vote affirmatively (Harmatuck 1973).
Demonstrated need for transportation improvements
Several researchers have found that voters would vote for higher taxes if there
was a demonstrable need for transportation improvements (Gomez-Ibanez
1999, Beale 1996). Specifically, Harmatuck (1973) found that those who were
inconvenienced by the lack of transit were very supportive of a transit tax.
Similarly, interviews with planners revealed their belief that severe congestion
in an area must be demonstrated in order to pass a transportation tax (Colman
1987). Colman (1987) also found that rapid population growth was an
important variable in demonstrating transportation needs. Accordingly, Beale
(1996) found that transportation taxes could be achieved only if there was a
clear need for road improvements.
Offer “something for everyone”
Several researchers found that the distribution of benefits was important in
passing transportation tax measures (Beale 1998, Colman 1987, Gomez-Ibanez
1999) In interviews with transit professionals, Middleton (1998) found that
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
25
when rail or bus was offered, it was important to include something of value to
non-users.
METHODOLOGY
Two regression models were developed to explain the percentage of voters in
favor of passage for local transportation taxes. Models were assembled with
variables that the existing literature, planners, or researchers have identified as
components of voting in efforts to pass local transportation tax measures. In
the first model, nine variables were examined to determine voting patterns in
cities and counties from the nationwide dataset consisting of 57 domestic city
and county elections from 1990 through 1998. In the second regression, a
comprehensive dataset of 63 California county elections that took place from
1980 through 1998 was reviewed. Five variables were examined with respect
to their potential effect upon percentage voting in those 63 counties.
Below, the datasets used for the nationwide and California analyses are
described, the variables are identified, and the rationale for choosing the
variables explained. The first subsection describes the nationwide sample,
while the second subsection focuses on the California sample.
National dataset
Sample
For the nationwide dataset, a sample of 57 U.S. city and county transportation
tax elections between the years 1990 and 1998 is used. The sample was
developed by identifying all city and county transit elections that reported the
percentage of voters who voted to pass transit measures. These data were
reported in the journal Passenger Transport between the years 1990 and 1998.
In approximately 85% (57 of 67) of the elections examined by the journal, the
percentage voting in favor of passage for transit measures is also reported. As
shown in Figure 2-3, the 57 elections represented 35 counties in 11 states, and
19 cities in 9 states.
There are some inherent limitations to data extracted from a sample of this
type. Most importantly, all elections reported contained at least some element
of transit, whereas in the California sample, some elections contained no
transit component. Furthermore, smaller cities and counties may not be as
motivated as others to report election results and so may not be included in the
resulting dataset. In addition, several states were potentially over-represented
and some states not included at all, and so results may represent those states
selected rather than the nation as a whole.
The managing editor of Passenger Transport has indicated that this
publication has a history of reporting election results and does remind planning
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
26
The Impact of Community Characteristics on Transportation Funding Outcomes
organizations to report results, but does not single out any specific cities or
counties to obtain information on election results.
Variable data
The analysis uses nine public choice predictor variables described in the
literature and considered by planners to contribute to the outcome of
transportation tax measures. These variables are included in six general
categories: age, income, tax environment, automobile-dependent population,
demonstrated need for improvements, and whether benefits are spread across
the general population.
Table 2-1. Dependent Variable
Category
Analyzed
Ballot
Outcome
Variable
MARGINFOR
Variable Description
% voting for passage of a transit tax
(The percentage voting for passage
was chosen for two reasons. One is
that it measures voting behavior
more effectively and with more
variance than do pass/fail data. In
addition, states have differing voting
requirements for passage, including
threshold percentages which must be
reached to enable measures to pass.
The percentage voting for transit tax
elections was obtained from reports
to the journal Passenger Transport
between the years 1990 and 1998.)
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
Table 2-2. Independent Variables
Category
Analyzed
Variable
Variable Description
Age
ELDERLY
% greater than 65 years of age was
obtained from 1990 Census data.
Income
INCOME
Per capita income was obtained from
1990 Census data.
Taxes
TAXBURDEN
Per capita taxes/per capita income
was calculated from 1990 Census
data.
Automobile
Dependent
Population
AUTO
% driving to work by automobile
was obtained from 1990 Census data.
Demonstrated Need
HOUSING
Number of housing units/population
was calculated from 1990 Census
data.
Commute
time
TRAVELTIME
Average number of minutes to
commute to work was obtained from
1990 Census data.
Population
POPULATION
Population estimates (1990) were
obtained from the National
Association of Counties.
http://www.naco.org/counties/.
Population
Change
POPCHANGE
Population change from 1980 to
1992 was calculated from 1990
Census data
Provide
Benefits for
All
BENEFITS
Continuous variable based upon the
number of modes described in the
ballot measure, from only one mode
(bus or rail or road) to several modes.
This variable was obtained from
election reports in Passenger
Transport.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
27
28
The Impact of Community Characteristics on Transportation Funding Outcomes
Table 2-3. Regression Specification for National Dataset
Variable
Variable Definition
MARGINFOR
β0+β1 (ELDERLY) +β2 (INCOME) + β3 (TAXBURDEN)
+β4 (AUTO) + β5 (HOUSING) +β6 (TRAVELTIME) +
β7 (POPULATION)+ β8 (POPCHANGE) + β9
(BENEFITS)
Table 2-4. Hypotheses for National Dataset
Variable
Predicted Effect on Voting
Predicted
Effect
ELDERLY
“ELDERLY is predicted to affect passage
negatively (B1<0). As the percentage of
elderly increases, the % voting for
passage decreases”
negative
INCOME
“INCOME is predicted to affect passage
negatively (B2<0). A per capita income
rises, the % voting for passage decreases”
negative
TAXBURDEN
“TAXBURDEN is predicted to affect
passage negatively (B3<0). As taxes/per
capita income rises, the % voting for
passage decreases”
negative
AUTO
“AUTO is predicted to affect passage
negatively (B4<0). As the % of
automobiles used to commute to work
increases, the % voting for passage
decreases “
negative
HOUSING
“HOUSING is predicted to affect passage
negatively (B5<0). As the number of
housing units/person increases, the %
voting for passage decreases”
negative
TRAVELTIME
“TRAVELTIME is predicted to affect
passage positively (B6<0). As the average
number of minutes to work increases, the
% voting for passage increases.”
positive
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
Variable
Predicted Effect on Voting
29
Predicted
Effect
POPULATION
“POPULATION is predicted to affect
passage positively (B7>0). As the
population increases, the % voting for
passage increases”
positive
POPCHANGE
“POPCHANGE is predicted to affect
passage positively (B8>0). As the
population change between 1980 and
1992 increases, the % voting for passage
increases”
positive
BENEFITS
“BENEFITS is predicted to affect passage
positively (B9 >0) when the ballot
measure describes funding for several
modes (transit and road), the % voting for
passage increases”
positive
California dataset
Sample
For the California dataset, a comprehensive sample of 63 county transportation
tax elections between the years 1980 and 1998 was used. Thirty-one California
counties, mostly urban, were included in the California dataset (transportation
tax measures on the ballot between 1980 and 1998) but the 27 counties, mostly
rural, were not included.
Variable data
Six public choice variables were identified using data that were available for
each of the predictor categories identified in the literature and considered by
planners to contribute to the outcome of transportation tax measures: age,
income, tax environment, tax, automobile-dependent population, demonstrated
need for improvements. Data were not available for the predictor category
“Benefits” (whether benefits are spread across the general population) and so it
was not included in the regression equation. In addition, the percentage
commuting by bus rather than the percentage commuting by car was used to
measure reliance on automobiles. Finally, population density was substituted
for housing density to measure the predictor category “Demonstrated Need.”
All other variables were identical to the variables measured for the national
dataset.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
30
The Impact of Community Characteristics on Transportation Funding Outcomes
Table 2-5. Dependent Variable
Category
Analyzed
Ballot
Outcome
Variable
Variable Description
MARGINFOR
% voting for passage of a transit tax:
This category was chosen for two
reasons. One is that it measures voting
behavior more specifically and with
more variance than do pass/fail data. In
addition, states have differing voting
requirements for passage, including
threshold percentages which must be
reached to enable measures to pass. The
percentage voting for transit tax
elections was obtained from a
compilation of data from Public Policy
Institute of California, Brown et al
(1998) & Todd Goldman (1999)
between the years 1980 and 1998.)
Table 2-6. Independent Variables
Category
Analyzed
Age
Variable
ELDERLY
Variable Description
% greater than 65 years of age: Data
for this category was obtained for
the years 1991 to 1998 from the
National Association of Counties
http://www.naco.org/counties/
counties/state. For the years 1980 to
1990, the proportion of the elderly
was estimated from USA Counties
data available at http://govinfo.
library.orst.edu/cgi-bin/usacostate?California
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
Category
Analyzed
Variable
Variable Description
Income
INCOME
Per capita income for year of
passage was obtained by year from
the Economic Census http://
govinfo.library.orst.edu/econstateis.html.
Types of Tax
Burden
LOCAL
TAXBURDEN
Per capita local taxes (including
property and sales and other local
taxes) was obtained from the State
of California’s Department of
Finance http://www.dof.ca.gov/
html/fs_data/profiles/pf_home.htm
SALES
TAXBURDEN
Per capita sales taxes within
counties was reported separately
from total local taxes, and obtained
from the State of California’s
Department of Finance http://
www.dof.ca.gov/html/fs_data/
profiles/pf_home.htm
Transit use
BUS
The percentage commuting to work
by transit was obtained from 1990
Census data.
Transit Use
continued
AUTO DENSITY
Auto Density was calculated from
the total number of auto and truck
registrations by county divided by
total miles of streets, roads, and
highways.
Demonstrated
Need
POP DENSITY
Population for the year in which the
measure passed/square miles for
county was calculated from http://
www.sco.ca.gov/govglance/9697/
gaagmap.htm
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
31
32
The Impact of Community Characteristics on Transportation Funding Outcomes
Category
Analyzed
Variable
Variable Description
POPULATION
Population estimates for July 1 in
the given year were obtained from
State of California, Department of
Finance data at http://
www.dof.ca.gov/html/Demograp/E6cover.htm.”
TRAVELTIME
Average number of minutes to
commute to work was obtained
from 1990 Census data.
Types Of
Population
Change
Variables
POPCHANGE5
Population change during the five
years prior to passage of measure
was calculated from State of
California, Department of Finance
data compiled for July 1 of each
year and available at Http://
www.dof.ca.gov/html/Demograp/E6cover.htm.
Population
Variables
continued
POPCHANGE2
Population change during the two
years prior to passage of measure
was calculated from State of
California, Department of Finance
data compiled for July 1 of each
year and available at http://
www.dof.ca.gov/html/Demograp/E6cover.htm
Table 2-7. Regression Specification for California Dataset
Variable
Variable Definition
MARGINFOR
β0+β1(ELDERLY)+β2(INCOME)+β3(TAXBURDEN)+
β4(BUS)+β5(AUTODENSITY)+β6(POPDENSITY)+β7
(POPULATION)+β8(TRAVELTIME)+β9
(POPCHANGE)
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
Table 2-8. Hypotheses for California Dataset
Variable
Predicted Effect on Voting
Predicted
Effect
ELDERLY
ELDERLY is predicted to affect
passage negatively (B1<0). As the
percentage of elderly increases, the
% voting for passage will decrease.
negative
INCOME
INCOME is predicted to affect
passage negatively (B2<0). A per
capita income rises, the % voting for
passage decreases.
negative
TAXBURDEN
TAXBURDEN measures are
predicted to affect passage negatively
(B3<0). As taxes/per capita income
rises, the % voting for passage
decreases.
negative
BUS
BUS is predicted to affect passage
positively (B4>0). As the %
commuting by transit increases, the
% voting for passage increases.
positive
AUTO DENSITY
AUTO DENSITY is predicted to
affect passage positively (B4>0). As
congestion increases, the % voting
for passage increases.
positive
POP DENSITY
POP DENSITY is predicted to affect
passage positively (B5>0). As the
number of housing units/person
increases, the % voting for passage
increases.
positive
POPULATION
POPULATION is predicted to affect
passage positively (B6>0). As the
population increases, the % voting
for passage increases.
positive
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
33
34
The Impact of Community Characteristics on Transportation Funding Outcomes
Variable
Predicted Effect on Voting
Predicted
Effect
TRAVELTIME
TRAVELTIME is predicted to affect
passage positively (B7>0). As the
average number of minutes to work
increases, the % voting for passage
also increases.
positive
POPCHANGE
Each of the POPCHANGE variables
is predicted to affect passage
positively (B8>0). As the population
levels prior to consideration of the
measure increases, the % voting for
passage increases.
positive
RESULTS
In these California data, the number of measures that passed was
approximately equal to the number that failed (30 passed vs. 33 failed). Table
2-9 includes population figures from the year that measures passed or failed
and does seem to demonstrate that counties that passed local transportation
measures had a larger mean population (1,211,290) during years of
transportation measure passage than did counties that failed to pass local
transportation measures (593,051.5). However, the larger average as well as
the mean difference was affected by an outlier, Los Angeles County. The
population of Los Angeles County (7,500,300 in 1980 and 8,897,500 in 1990)
was more than three times that of the next most populous county at each time
period demonstrated in these data. Moreover, Los Angeles County passed two
measures in 1980 and 1990 and so this outlying population was an important
component of mean and mean difference. Despite appearances, analysis of
variance indicates that mean population differences have not been statistically
associated with passage of transportation measures (F=2.787, p=.100).
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
35
Table 2-9. Average County Population for Measures that Passed/Failed
No. of
Measures
Ave. Population
Passed
30
1,211,290.00
Failed
33
593,051.50
Ave. Population
Density
2149.5
550.6
Source: State of California, Department of Finance, Historical County
Population Estimates and Components of Change, July 1, 1970-1990.
Sacramento, California, December 1998. http://www.dof.ca.gov/html/
Demograp/E-6cover.htm and State of California, Department of Finance,
Historical County Population Estimates and Components of Change,
July 1, 1990-1998. Sacramento, California, January 1999. http://
www.dof.ca.gov/html/Demograp/e-6text.htm.
Density estimates were obtained from the U.S. Bureau of the Census.
Counties with larger population densities, however, were more likely to pass
local transportation measures than counties with smaller population densities.
Counties that passed local transportation tax measures had an average
population density of 2,149.5 people per square mile during relevant election
years, while counties that failed to pass transportation measures had an average
population density of 550.6 people per square mile. These differences were
affected by the passage of three measures in one county in which population
density was five times as great as the next most dense county during the
relevant time periods: San Francisco County’s population density was
15,513.90 people per square mile at the time of the passage of its last two
transportation measures (see Table 2-9). However, despite the appearance of
this outlier, mean density (unlike population alone) is statistically associated
with the passage of transportation measures. Analysis of variance indicates that
counties that passed transportation measures are significantly more dense
(F=4.029, p=.049).
Regression results
The results from the analysis of California data are summarized in table 2-11.
Despite the initial appearance of relationships between population and
population density and the passage of transportation measures, analysis of
variance bears this out only in the case of the population density variable.
Population density is also very strongly and significantly related to the
percentage using public transit (BUS) (r=.93, p<.0001). Since population
density is available by year, but percentage using public transit is available less
often, population density is used in the regression.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
36
The Impact of Community Characteristics on Transportation Funding Outcomes
Regression results indicate that a model utilizing population density for the
year of the measure, proportion elderly for the year of the measure, proportion
of population change for the five years prior to the measure, and sales tax per
capita, as independent variables can explain 27 percent of the variance in
margin voting for transportation measures (R2=.27). Density, proportion
elderly, and population change over the previous five years were positively
associated with the proportion voting for passage of transportation measures.
Sales tax per capita was negatively related to proportion voting for passage. As
population density, proportion elderly, and population change increase, so does
proportion voting for transportation measures. However, lower sales taxes per
capita are associated with higher proportions voting for transportation
measures.
Table 2-10. Model: Dependent Variable = Proportion Voting for
Transportation Measure
B
Constant
Std.
Error
17.347
8.866
Population
Density
3.056
0.001
Proportion
Elderly
2.403
Population
Change
Sales Tax
Per Capita
Beta
t
Significance
1.956
0.055
0.695
2.478
0.016
0.747
0.426
3.215
0.002
93.155
33.01
0.433
2.822
0.007
-0.238
0.129
-0.486
-1.844
0.070
It is important to note that one variable was responsible for more than half of
this variance: fully 15.8 percent of the explained variance can be accounted for
by the variable that represents proportion of elderly. In a bivariate regression
utilizing the proportion of elderly in each county as the lone independent
variable, results were R2=.158 (p<.001), meaning that the proportion of
individuals over 65 in each county was significantly related to the percentage
voting for the passage of transportation taxes.
MARFOR=28.4 + .398 ELDERLY
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
37
As the percentage of elderly in a community increased, the percentage of the
community voting for passage of a transportation tax received a significant
boost. When the elderly population in a community reached a threshold of 13
to 15 percent, the mean percentage of the community voting for passage of the
tax averaged 55.3 percent; while in communities where the elderly population
was greater than 15 percent the average proportion voting for transportation
measures was 72.8 percent. In communities where the population of elderly
was 9 percent or less, the proportion voting for transportation measures was
less than a simple majority, 48.6 percent. Analysis of variance confirms that
proportion voting for passage significantly varies by the proportion of elderly
in the community (F=3.838, p<.01).
Interestingly, if population change in the five years prior to the measure is
inserted into the simple regression equation with proportion elderly, these two
independent variables predict 19 percent of the margin voting for the measure
(the R2 is increased marginally to .19 (p<.003)), indicating that public reaction
to changes in population is a small but significant addition to perception
regarding the importance of public transportation.
Table 2-11. Summary of results: California Data
Variable
Predicted
Effect
Observed
Effect
ELDERLY
negative
positive
INCOME
negative
None
TAXBURDEN
negative
negative
BUS
positive
None
AUTO DENSITY
positive
None
POP DENSITY
positive
positive
POPULATION
positive
None
TRAVELTIME
positive
None
POPCHANGE
positive
positive
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
38
The Impact of Community Characteristics on Transportation Funding Outcomes
Analysis of results
These results show that some variables commonly thought to affect the
outcome of local transportation tax measures, including mean population,
income, automobile density, use of public transit, and average travel time to
work, do not predict voting behavior. However, a model containing population
density, proportion elderly, population change in the five years prior to the
measure, and sales tax per capita, does explain variance in the outcome of
transportation taxes that are presented for a public vote. Each of these variables
is of interest. Overall, however, they indicate a perception of need that is
affected by evidence of increases in population density and an existing sales
tax burden that is comparatively large, precisely the type of tax that
transportation measures usually target.
The single variable of most interest is the proportion of elderly citizens. This
variable appears to have the strongest effect in predicting the margin voting for
local transportation tax measures in these data. The negative relationship that
has been presented in some of the literature was not found, instead there is a
strong positive relationship between percentage elderly and percentage voting
for passage of transportation tax measures. Explanations for this outcome
include a) the propensity of the elderly to turn out to vote in both on and offterm elections, and b) patterns of need for, and usage of, public transit among
the elderly. The need for transit among this group seems to be associated with a
willingness to bear additional taxes. Their significant turnout at the polls can
create a critical mass in elections where they represent larger proportions of the
overall population.
National dataset
Overview of national cities and counties
The national dataset includes selected demographic and ballot information for
the sample U.S. cities and counties including: 1990 population; population
density as measured by number of housing units per capita; change in
population between 1980 and 1992; population over 65 years of age; per capita
income, tax burden (per capita taxes/per capita income); per capita income;
percent commuting to work by car; average travel time to work; year of ballot
measure; and the passage or failure of the ballot measure.
Table 2-12 demonstrates that overall the number of measures that passed was
greater than the number that failed (32 passed vs. 25 failed). In addition,
counties with larger populations were more only slightly more likely to pass
local transportation measures than counties with smaller populations.
Specifically, those counties that passed measures had average populations of
728,428 people whereas counties that failed measures had average populations
of 633,949 people.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
39
Table 2-12. Average County Population for Measures that Passed/Failed
No. of Measures
Ave. Population
Housing Units
Per Capita
Passed
32
728,428
0.41
Failed
25
633,949
0.41
Source: U.S. Census Bureau, 1990 and
http://www.sco.ca.gov govglance/ 9697/htm/allstate/population.htm
Additionally, counties that passed measures did not differ in terms of housing
per capita from counties with measures that failed. Specifically, counties that
passed local transportation tax measures had an average of .41 houses per
capita and similarly, counties that failed transportation measures had an
average of .41 houses per capita.
Regression results (national dataset)
Findings from the analysis of national data are summarized below in table 213. Despite the initial appearance of a relationship between population or
population density and the proportion of the population that voted for
transportation measures, no evidence of such a relationship was found when
the regression model that predicted the percentage voting for passage was
examined.
Variables in the regression equation were examined for collinearity and none
of the variables considered demonstrated evidence of high collinearity (defined
as r>.8). Therefore all variables were included in the regression model. A
stepwise regression model explained 7 percent of the variance in voting for a
measure (Adjusted R2=.076, p=.044). However, only two of the variables
significantly predicted the percentage voting for transportation taxes.
Specifically, as found for the California dataset, the variable ELDERLY
significantly predicted the percentage voting for the passage of transportation
tax (p=.029). In addition, the variable BENEFITS significantly predicted the
percentage voting for the passage of transportation taxes (p=.075).
MARFOR=68.7 – 1.1 ELDERLY – 7.2 BENEFITS
Thus, as the percentage of elderly in a community increased, the percentage of
the community voting for passage of a transportation tax decreased.
Additionally, if measures included both transit and road projects rather than
just transit or just road, the percentage voting for passage decreased.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
40
The Impact of Community Characteristics on Transportation Funding Outcomes
Table 2-13. Summary of Results: National Data
Variable
Predicted Effect
Observed Effect
ELDERLY
negative
negative
INCOME
negative
None
TAXBURDEN
negative
None
AUTO
negative
None
HOUSING
negative
None
TRAVELTIME
positive
None
POPULATION
positive
None
POPCHANGE
positive
None
BENEFITS
positive
negative
Analysis of the national dataset
These results showed that variables commonly thought to affect the outcome
of local transportation tax measures, including income, tax burden, automobile
dependency and demonstrated need for transportation improvements measured
by population density, population change, and average travel time to work,
predict voting behavior on local transportation tax measures imprecisely. As
was the case for the California dataset, age appears to have had an important
effect in predicting voting for local transportation tax measures. However,
unlike the California dataset, a weak negative relationship between percentage
elderly and voting for passage of local transportation taxes was found, as
predicted.
In communities where the population of elderly was greater than 18 percent, 67
percent of the community voted to pass the transportation tax. Whereas, in
communities where the population of elderly was between 6 and 18 percent,
only 51 to 53 percent of the community voted to pass the transportation tax.
These results are consistent with the results for the California dataset.
However, what was unusual about the results for the national dataset was that
for communities where the population of elderly was less than 6 percent, the
mean percentage voting to pass the transportation tax was 71 percent, higher
than in any other age category.
Additionally, as the percentage of measures that included both transit and road
projects increased, the percentage voting for passage of transit taxes decreased,
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
41
which was also contrary to the original prediction. For communities that
describe multiple modes of transportation on the ballot, the mean percentage
voting for passage of transportation tax measures was 51 percent. However, for
communities that offered only one mode of transportation, the percentage of
the community voting for passage of the tax was 56 percent.
ANALYSIS OF DATASETS
Analysis of ELDERLY variable
The significance of the ELDERLY variable is particularly interesting in the
national level data because of the apparent contradictory effects of proportion
elderly in a community on percentage voting for transportation taxes. While
the proportion elderly in California counties was positively related to
proportion voting for passage of transportation taxes, at the national level a
negative relationship was observed between percent elderly and percent voting
for passage of transportation taxes.
This discrepancy may be due in part to differences in the range of percent
elderly residing in communities. In the California dataset, the lowest
proportion of elderly residing in any county was 9 percent, while in the
national dataset, the lowest proportion of elderly residing in any city or county
was less than 6 percent. Furthermore, when results in communities that had
greater than 9 percent elderly are compared, there was much less discrepancy
between the California dataset and the National level dataset. Specifically,
when all communities that had elderly populations of less than 6 percent (e.g.,
the cities of Carrolton, TX; Garland, TX; Irving, TX; Plano, TX; and Lansing,
MI) were removed, the variable ELDERLY no longer predicted the percentage
voting for passage of transportation taxes. Although this result is not consistent
with the California dataset, it is no longer contrary to it. As a result, there may
be significant differences between cities and counties in predicting the effect of
percent elderly in a community on percent voting to pass local transportation
taxes.
The California findings are not consistent with findings by Colman (1987) and
Harmatuck (1973), who predicted that younger rather than older voters may be
more willing to vote for transportation taxes. In addition, although the national
dataset results are consistent with previous findings, by removing the cities
described above, that national dataset results are not consistent with previous
findings. Furthermore, these findings do not support Harmatuck’s conclusion
that because transportation benefits may take years to be realized, older voters,
for whom transportation taxes may extend beyond their own may be less likely
to vote for their passage. Instead, in communities where the population of
elderly was larger, voters in California tended to vote for the passage of
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
42
The Impact of Community Characteristics on Transportation Funding Outcomes
transportation taxes. These results are consistent with Levine (1998) who
found that residents in a community in Michigan were likely to vote for transit
tax measures primarily for the benefit they perceived for disadvantaged groups
such as the elderly. In those communities that have a large percentage of
elderly, voters may be more aware of the needs of the elderly and may be more
willing to vote for transportation tax measures in part to provide benefits to the
elderly.
Analysis of BENEFITS variable
Communities that fielded measures designed to provide multiple benefits
(transit and road) were less likely to vote for the passage of transportation taxes
than communities that fielded measures designed to provide only one benefit
(transit). These results are contrary to findings reported by Beale (1998),
Colman (1987), and Gomez-Ibanez (1999) who concluded that the distribution
of benefits was important in passing transportation tax measures. Specifically,
the results are contrary to Middleton (1998) who found that when rail or bus
was offered, it was important to include something of value to non-users.
Instead, in those communities that offered only transit benefits, residents were
more likely to vote to pass the transportation tax. Furthermore, in those
communities that offered both transit and road (Los Angeles, Santa Clara,
Sonoma, and Ventura Counties), residents were less likely to pass the tax.
Two explanations are offered for these results. First, all of the counties that
offered both transit and road benefits were California counties. California
counties may have more difficulty than other counties in the nation passing
transportation taxes, whether or not they fund multiple types of transportation
projects. Second, the results, may be due to the procedure used to obtain
information about transportation projects that would be funded with the
transportation tax. Information was obtained about projects to be funded from
election results reported in Passenger Transport journal. This journal relied on
reports from transportation professionals to describe the ballot measure, so if
the professionals did not believe that a significant component of the funding
would fund road projects, they did not mention it.
Implications
Conventional wisdom suggesting that communities with greater percentages of
elderly may be less supportive of transportation taxes than communities with
lower percentages of elderly may be incorrect. Instead, communities with
greater percentage of elderly may be more willing to support taxes for
transportation than communities with a lower percentage of elderly.
In addition, conventional wisdom suggesting that transportation plans should
include benefits for everyone may be incorrect. Instead, communities that have
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
43
plans to fund only transit may be just as, and perhaps more, acceptable to the
public as plans that fund both transit and road projects.
Based on these implications, two finding seem plausible.
Finding 1:
Efforts to fund transportation with taxes where the proportion of elderly is
greater than 9 percent are more likely to succeed
In communities where the percentage of elderly is greater than 9 percent, the
findings suggest that the community may be more willing to accept local
transportation taxes. However, in communities where the percentage of elderly
is less than 9 percent, transportation measures will require significantly more
determined marketing to enhance the probability of passage.
Finding 2:
Efforts to increase sales taxes for transportation programs will be less
successful in communities with higher sales taxes.
The relatively strong and negative relationship between sales tax and support
for transportation tax initiatives suggests that communities with relatively
higher sales taxes will be hard pressed to convince citizens to support
additional increases.
DIRECTIONS FOR FUTURE RESEARCH
The need for future research is indicated by many of the findings in these
analyses. For instance, results indicate that planners and politicians prefer
certain funding mechanisms for transportation projects. As shown in Figure 21, the sales tax is the preferred method of financing in many communities
across the nation. Reflecting this preference, 66% of tax measures presented to
voters have included a sales tax funding mechanism, while only 33% used the
property tax, and less than 2% required a bond or payroll tax.
However, the sales tax was less effective in appealing to voters. As shown in
Figure 2-2, the sales tax passed just under 50% of the time, whereas property
tax measures passed just over 75% of the time. Additionally, in the California
data, sales tax per capita was inversely related to proportion voting for
transportation measures, indicating an unwillingness to bear more of this
regressive tax if the burden is already comparatively large. For these reasons,
the popularity of the sales tax in transportation measures among the crafters of
transportation measures warrants further analysis.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
44
The Impact of Community Characteristics on Transportation Funding Outcomes
Figure 2-1. Type of Funding Mechanism on Transportation Tax Measures
in U.S.
70
60
50
40
30
Percent
20
10
0
property tax
sales tax
bond
payroll tax
funding mechanism
Figure 2-2. Percentage Voting for Transportation as a Function of
Funding Mechanism.
40
30
20
Count
10
Pass/Fail
pass
fail
0
property tax
sales tax
bond
payroll tax
funding mechanism
Another interesting and somewhat counterintuitive finding concerns the effect
of population change on the passage of transportation taxes. It might be
expected that if a community’s population grows substantially, its residents
would be more willing to pass transportation tax increases to support the
greater population size. However, as shown in Figure 2-3, in a subset of 38
counties for which population change data could be obtained, the slower a
community’s population growth between 1990-1996, the more likely it was
that a transportation tax was passed. Communities with negative or very slow
population growth (less than 5%) passed transportation tax measures
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
The Impact of Community Characteristics on Transportation Funding Outcomes
45
approximately 70% of the time. Whereas, communities with moderate
population growth (5-13%) passed transportation tax measures about 50% of
the time. Communities with high population growth (13-17%) passed
transportation tax measures only 25% of the time. Communities with
extremely high population growth (greater than 17%) passed transportation tax
measures about 60% of the time. Preliminary conclusions from these results
suggest that communities with little population growth may be able to justify
transportation funding to voters based on the implied or explicit justification
that improved transportation may improve their community’s economic
growth. Alternatively, communities who are growing slowly or not at all, may
have a concerted public opinion that promotes slow growth—this type of
community may have attendant support for measures designed to clean up or to
keep the environment clean.
Figure 2-3. Percentage Voting for Transportation as a Function of
Population Growth
12
10
8
6
4
Count
Pass/Fail
2
pass
0
fail
Less
than
1%
14.9%
58.9%
912.9%
1316.9%
Greater
than
17%
Population growth - Grouped
Communities experiencing moderate growth may be experimenting for the
first time with alternatives to traditional highway financing. These
communities may find that voters are skeptical of new modes of transportation
and new methods of paying for these changes. Those communities with
exceptionally high population growth may be more successful in marketing
rail to voters because they may already have rail in place. Explanations for this
divergence between communities may be examined in more detail as the
research continues.
A final observation from this preliminary research was that transportation tax
measures were more likely to pass in communities where a lower proportion of
people drove alone to work. As shown in Figure 2-4, in communities where
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
46
The Impact of Community Characteristics on Transportation Funding Outcomes
less than 80% of the population drove alone to work, transportation tax
measures were supported approximately 66% of the time. Whereas, in
communities where more than 80% of the population drove alone to work,
transportation tax measures were supported less than 33% of the time. This
result suggests that communities in which the great majority of residents drive
alone to work do not view transportation taxes as beneficial or perhaps even
meaningful to them. Since none of the communities within which more than
80% of the population drove alone to work listed road projects as a significant
part of their transportation plan, these populations might also see those
transportation measures that have been presented to them as superfluous,
simply a tax to pay for something they have no intention of using.
Percent Voting For a Transportation Tax Increase
Figure 2-4. Percentage Voting for Transportation as a Function of
Driving Alone
70
60
50
40
30
20
10
0
less than 80 %
greater than 80%
Percentage of Population Driving Alone
As communities
accept
greater
responsibility for funding their own transportation projects, it becomes even
more important to determine which factors influence community support for
transportation tax measures. These results can begin to provide a window into
the public mind with information that uncovers the way in which funding
mechanisms, population growth, and travel behavior influence outcomes in
local efforts to fund transportation costs. However, the findings presented here
are of limited utility because, with the exception of sales tax burden for
California, they fail to point to variables that can be adjusted by policies that
decision makers can influence to help ensure the passage of such measures.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
47
CASE STUDIES: CAMPAIGNS FOR
TRANSPORTATION TAX MEASURES IN
FOUR COMMUNITIES
INTRODUCTION
As an initial foray into explaining the relative success of transportation tax
measures, one major conclusion from the preceding quantitative analysis,
albeit an extremely tentative one, is that although community level
characteristics play at least a modest role shaping electoral outcomes, much is
left to explain. Assuming that these results stand up in the face of further
scrutiny, two explanations are immediately plausible for the lack of extremely
strong relationships between electoral outcomes of tax measures and
community characteristics.
First, it is possible that other community level characteristics exist that were
excluded from the preceding analysis. Such variables might ultimately prove to
exhibit stronger relationships than those analyzed for this study. For example,
a more accurate measure of traffic congestion, or population growth might do a
better job of predicting the outcome of transportation tax measure ballots.
Future research should address this possibility.
A second possibility is that entirely different kinds of variables, those linked to
the specific strategies, tactics, and other efforts of transportation tax measure
citizens might be more able to affect the outcome of these elections. However,
modeling the potential effects of such efforts is not practical. No known dataset
exists that isolates, categorizes, and provides accurate measures of the efforts
of local tax campaigns. Indeed, relatively little is known about how such
campaigns are planned and waged.
PRIOR RESEARCH ON VOTER INFLUENCES PERTAINING TO
LOCAL TRANSPORTATION FUNDING INITIATIVES
The research reported in this chapter examines how voting results in local
elections involving transportation funding initiatives are influenced by the
extent and nature of both supporting and opposing coalitions, the nature of the
transportation package involved with the funding initiative, and the marketing
and communications campaigns used by both proponents and opponents.
Methodology of prior research
Prior research on this topic is not extensive. Both the methodology and results
of the research identified are presented in this section. Two prior studies were
identified which used a case-type study approach. Beale, Bishop, and Marley
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
48
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
(1996) conducted site interviews involving 22 cases. Additionally, telephone
interviewing and unspecified documentation was used with approximately 90
additional cases. Cases in 11 states were included, with at least 75 percent of
the cases occurring in the four states of California, Florida, Georgia, and
Missouri. The cases discussed in this research comprise a significant amount of
variance with respect to both the size and the nature of the transportation
projects studied. Many of the proposals do not have a transit component. Most
of the elections referred to in these cases appear to have taken place in the
latter half of the 1980s and the first half of the 1990s. Beale et al did not seem
to have collected a significant amount of information about the marketing and
communications efforts used by proponents and opponents.
Middleton (1998 G2) “talked with several experienced transit professionals
who have considerable expertise in developing, and winning, public support
for transit.” The recommendations from this research are based strictly on the
insights of these transit professionals who have been involved in trying to
obtain voter approval in their jurisdictions. Thus, the method used was
essentially a quasi-case approach.
A few studies were identified that use surveys to identify factors influencing
public support for transit (Forkenbrock and Stoner 1983; Trent and Bernard
1985; Levine, Park, and Wallace 1998). They have limited value in
determining how to develop a successful transportation package or how to
market that package to obtain voter acceptance because the dependent
variables used have limited generalizability.
For example, Levine et al. used potential voters’ support for a property tax
earmarked for a paratransit system as a dependent variable. One of the
independent variables examined was labeled a “social service” variable,
measured by agreement with statements such as “public transit is a needed
social service.” Another independent variable, labeled as a “transit
environment” variable, was measured by agreement with such statements as
“public transit helps reduce road congestion” and “public transit helps reduce
air pollution.” The analysis discovered that the social service variable was
more highly correlated with voter intentions than was the transit environment
variable. This finding is intuitively consistent with voting intentions towards
paratransit since this mode of transportation basically is a social service rather
than a mode that potentially reduces congestion or air pollution. However, it is
likely that the importance of these two independent variables would be
reversed if voter intentions towards the combination of passenger rail or
additional high occupancy vehicle (HOV) dedicated highway lanes were being
measured. In such a case, the main potential benefit of this transportation
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
49
package could be a reduction in congestion and air pollution rather than a
social service.
Since these survey-based studies have limited generalizability, the results of
these studies will not be discussed in the section that immediately follows.
RECOMMENDATIONS FOR OBTAINING VOTER APPROVAL OF
TRANSPORTATION FUNDING INITIATIVES
Both Beale et al. (1996) and Middleton (1998) present a number of
recommendations to increase the probability of voter approval for
transportation funding measures. Their recommendations are listed below,
grouped into three categories:
1) The process of building coalitions and developing a transportation
package
• Involve the public in the planning process through open meetings,
surveys, and or focus groups (Beale et al. Middleton).
• Involve relevant representatives of various groups who could
potentially support or oppose transportation packages. The groups
below were identified by Middleton or Beale et al.
•
Business community and Chamber of Commerce (Beale et al.)
•
Representatives on the governmental agency voting to present
the funding package to the public and other elected officials,
such as mayors (Middleton and Beale et al.)
•
Environmentalists (Beale et al.)
•
Developers and automobile dealers (Beale et al.)
•
Taxpayer associations (Beale et al.)
•The elderly and handicapped (Beale et al.)
2) The transportation package
• Funding should be specifically earmarked
for
specific
transportation projects (Beale et al., Middleton).
• “Provide something for everyone,” (Middleton) by advancing a
package that combines highway improvements with transit. Beale
et al. also discuss this issue but present a conditional
recommendation, indicating that this decision should depend on
local conditions.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
50
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
• Fund “projects throughout the area” (Beale et al.) or “provide subarea equity” (Middleton).
• The cost of the package should be inexpensive enough to have both
an acceptable tax level and sunset period (Beale et al.). The authors
give an example of .005 being a threshold level for a sales tax and
suggest that a sunset date of 10 years or even 2 to 5 years may be
needed.
• Leverage state and federal state funds into the funding package
(Beale et al.).
• Because additional funding measures will need to be approved in
the future, realistic costs, time frames, and outcomes should be
developed. In other words, the success of future funding measures
depends upon successfully meeting expectations and predictions
with the current package (Beale et al.).
3) The marketing and communications campaign
• The campaign should not be run directly by transit officials with a
vested interest in the outcome (Middleton).
• The timing of the election must be considered carefully (Beale et
al). Middleton recommends avoiding recessions and elections with
other funding measures.
• The public should be educated over an extended period through
involvement in an open planning process (Beale et al.); detailed
information of all the projects and their costs ought to be presented
prior to the election (Middleton).
• Open communications with the public should continue after a
successful election, assuming that additional funding measures will
need to be approved in the future. Post-election communications
should include realistic measures of progress against performance
measures (Middleton).
• The benefits of transit to transit non-users ought to be demonstrated
(Middleton).
• Ballot language should be carefully formulated, using specific and
informative rather than vague and legalistic language (Beale et al.).
• Benefits should be formulated and articulated early in the process
and taxes mentioned as little as possible.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
51
EVALUATION OF RECOMMENDATIONS FOR OBTAINING VOTER
APPROVAL OF TRANSPORTATION FUNDING INITIATIVES
The above recommendations provide a reasonable initial framework for those
attempting to obtain voter approval of transportation funding initiatives. At the
same time, however, modifications and additions to these recommendations
can be made.
Some of the above recommendations could be more specific. In fact, some of
them are goals rather than strategies and tactics for achieving goals. For
example, both Middleton and Beale et al. identify groups that should be
included in a coalition. However, the means of building a coalition of groups
that have different, and even conflicting, perspectives towards transportation
packages and funding mechanisms are not discussed in any depth.
Furthermore, both Middleton and Beale et al. identify the importance of
identifying transportation packages that are not overly costly. However, basic
strategies for developing these transportation packages are not identified, nor
are means of prioritizing components of these packages.
Modifying the above recommendations to focus on strategies and tactics for
achieving the goals stated in the recommendations reveals several difficult
issues and challenges. For example, it is a major challenge to develop a
transportation package that is not too expensive for voters while at the same
time providing geographical equity and also satisfying the demands of both
automobile and transit advocates.
Although prior research has identified recommendations with some relevant
components of a marketing plan, other critical components have not been
identified. For example, no recommendations were made about how to respond
to opponents arguments, nor has much been said about how different
information sources are used to present a combination of both general themes
and specific information on some rather complex issues. One such complex
issue involves describing the specific components of a large transportation
package.
To summarize, the prior case research has tended to emphasize breadth rather
than depth, both in the nature of the cases studied and with the amount of
information gathered with each case. This broad approach is appropriate at the
onset of research on a topic.
Methodology
Now that prior research on local transportation funding initiatives has
developed a broad framework, a more in-depth case study approach is
appropriate. Perhaps only by examining interviews with individuals involved
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
52
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
with specific transportation tax measures can an understanding be gained of
how their efforts potentially influence the election outcomes. The case study
approach adopted for this study provides depth, which complements the broad
approach taken in prior research. Such an approach has two characteristics.
First, the cases used in this research have a more specific focus that those used
in prior research. Each of the cases included in this report has an estimated cost
of at least $1 billion and includes a passenger rail component. In addition,
since developing a systematic understanding of the marketing campaigns used
by both proponents and opponents is an important goal of this research project,
cases have been selected in which opponents or proponents actually used such
campaigns. For example, proponents spend at least $400,000 on marketing in
each of the cases studied.
Second, a significant amount of information was collected for each case. Onsite interviews with a number of people representing a variety of perspectives
were conducted. In addition to these interviews, relevant studies were
examined as were newspaper articles and editorials. Post-election surveys were
conducted in both of the unsuccessful cases and these results have also
provided important insights.
However, the use of case study methods implies several weaknesses from the
standpoint of analytic rigor. Most notably, inclusion of a sufficient number of
case study sites to create a random probability sample that would generate
robust findings is frequently impossible due to resource constraints. Case study
data, which typically include interviews with key respondents, as well as
observation and document review, may generate misleading findings based on
an insufficient number of observations.
Nevertheless, given the overall lack of definitive information about how
transportation tax measure campaigns work, four case studies were conducted
in support of this project. Four communities with recent transportation tax
measure elections were selected. Resource constraints precluded a random
sampling approach; instead sites were selected with a purposive sampling
strategy. Purposive sampling is used to develop a list of cases that will generate
useful qualitative insights by means of cases that share characteristics of
interest to the researcher. In this case, sites were selected purposively based on
the following criteria:
1) Successes and failures. Two of the cases involve measures that
passed, and two concern those that failed.
2) Range of transportation proposals. The cases represent a fairly wide
range in both the cost of the transportation proposal and the mix of
highway to transit improvements. At the same time, the cost of each
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
53
proposal was greater than $900 million and each proposal had a
significant passenger rail component.
3) California representation. Because of the special circumstances
involved with passing tax initiatives in California (particularly the
supermajority requirement for most earmarked tax increases), as
well as the pervasive traffic problems that plague California cities,
two cases were drawn from the “Golden State.”
4) Jurisdiction variety. The cases represent a variety of jurisdictions
and settings in which to investigate the campaign process, including
both intensely urban and more rural communities as well as
counties and regional jurisdictions.
5) Formal marketing and communications efforts by proponents.
Proponents spent at least $400 thousand during the campaign. The
cases represent some range in opponents’ campaign efforts.
Organized opposition existed in three of the four cases.
The case study site visits typically consisted of the following activities:
1) On-site visit and telephone interviews with key officials and
individuals including: sponsors and opponents of transportation tax
measures; campaign consultants and staff; members of local
business and environmental communities and other interested
parties; and elected officials and other individuals with insight into
the campaign process.
2) Review of relevant documents, and other documentation of
outcomes associated with transit coalitions, including newspaper
coverage.
Each site visit was used to generate: an overview of community's
demographics, fiscal and economic background, and relevant historical and
political background; identification of major transportation challenges and
problems; description of strategies, techniques, approaches, etc., that were
used to achieve (or failed to achieve) public support; and evaluation of
effectiveness of each effort.
CASE STUDY I
Santa Clara County Measures A & B (1996)
Santa Clara County, which comprises the heart of “Silicon Valley,” is an area
long renowned for its traffic and congestion. In November of 1996, the county
voters narrowly approved a nine-year, $1 billion sales tax. Proceeds from the
tax are being used to fund a variety of transportation improvements in the
county, including road widening and repair, traffic light synchronization, and
public transportation, including added light rail service.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
54
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
The plan to significantly boost funding for public transportation and the other
improvements was complicated, due to the state’s constitutional requirement
that specific tax increases be approved by super-majorities of two-thirds or
greater. Only by linking a more generalized tax increase to an advisory
measure, a move likely to confuse or alienate many voters, could such an
increase avoid this requirement. This case study illustrates the impact of
successfully building a pro-transportation coalition on a successful campaign
to create transportation funding. It also provides specific examples of
marketing research and related strategies that appeared to help the measure
pass. However, the fact that the measure ultimately passed by a thin margin
despite a paucity of organized opposition serves to accentuate how difficult it
is to gain public acceptance of tax increases to support transportation.
Among the individuals interviewed for this case were: members of the county
board of supervisors; the campaign director and lead consultants for the
proponent’s campaign; and staff from the local transportation agency.
Background: transportation and traffic
As the geographic center of the Silicon Valley, Santa Clara County has
experienced longstanding and continually worsening problems with traffic and
congestion. As the technology sector has mushroomed with employment
opportunities, the construction of housing has lagged, particularly housing
within easy reach of Silicon Valley employment centers. Many commuters
face drives of greater than one hour each way to their places of employment.
Public opinion surveys have repeatedly indicated that residents consider traffic
and transportation problems to be among the important, and frequently the
most important issues facing the area.
Before the 1996 measure came up for consideration, the county, with federal
assistance, had already funded and constructed a light rail line, essentially a
single north-south commuter line. The county transportation agency also
operated an extensive county bus service. Commuter rail service also existed,
transporting commuters north and south along the peninsula to San Francisco.
Several freeways were contained HOV lanes. However, at the time of the 1996
proposal, a recent upturn in the high tech industry had fueled even more
economic expansion, employment, and therefore traffic. Frustration with
traffic and congestion had reached the point where many residents were
sympathetic to a public response to the problems.
Santa Clara County has exhibited support for past transportation-related tax
initiatives. In 1976, a half-cent sales tax, open-ended, with no expiration date,
for transit projects was passed by county voters. The tax increase was
earmarked for the construction of a light rail system. The measure received
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
55
56% of the vote in a special election in which only 17% of registered voters
participated. The board of supervisors put this item on the ballot. The
campaign was led by a member of the board, who helped to form a coalition
consisting of the League of Women Voters, builders, organized labor,
environmentalists, and business leaders. Forty thousand dollars was spent to
help promote its passage.
Passage of the 1976 tax was preceded in 1975 with a series of public hearings
about light rail. A master plan for transportation was proposed and this plan
required a public vote of approval every four years, although it did not really
bind any public action. However, each vote created an opportunity for
publicity about the transit projects included in the plan. Consultants were hired
to run the public meetings that discussed the master plan. Since that time, these
plans have passed by a 70 percent or greater margin. Transit supporters on the
board of supervisors believe that the plans helped to get people used to
thinking about, and voting for, transportation initiatives. They also consider the
public participation generated by this process to be an important source of
input for planning and packaging transportation-related initiatives.
Again in 1984, a transportation tax initiative, Measure A, was placed on the
county ballot. It authorized an additional half cent sales tax that would be used
to improve key local highways 101 and 237, and to build the long envisioned
Highway 85. This tax had a 10-year life span. The measure was proposed in
response to federal and state funding cuts and to appease supporters of
highway construction. A county supervisor, Zoe Lofgren, led campaign for this
measure, which passed easily.
In 1992, with the old (1984) Measure A due to expire, a Measure A extension
was placed on the ballot. It passed with a 55 percent majority. It would have
created a 20-year tax (also a one half-cent on sales), of which 80 percent would
have been earmarked for transit. Opponents of the tax successfully sued,
claiming that the tax violated the state’s constitutional requirement for a
supermajority for specific tax increases.
In 1996, a new strategy for packaging transportation measures on local ballots
was devised and initiated. Based on the so-called “Mill Valley” model, the
strategy involves passing two measures, one of which is a non-binding plan for
spending increased tax revenues, the other a general sales tax increase. This
approach makes transportation tax increases easier to pass, at least in theory,
because general tax increases require only a normal majority of votes. This
measure was successfully passed, along with the non-binding spending plan,
and survived a court challenge.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
56
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
The tax had a 9-year life span, and earmarked 60 percent for transit and 40
percent for highway construction. Measure A described the spending plan,
Measure B included the tax itself. The campaign to pass these measures cost $1
million, of which 80 percent came from the Silicon Valley Manufacturers
Group (SVMG). (The SVMG and its role in this case is described more
completely in the following section.) The funding ratio (60/40) was a subject
of compromise between those who advocated transit and the SVMG.
Background: political history
The Santa Clara County Board of Supervisors is the legislative decision
making body with responsibility for transportation policy in the county. The
Valley Transit Authority (VTA) is the County’s branch of Caltrans, the
California State transportation agency. It is the executive agency charged with
implementing and managing most transportation policies. Its board of directors
consists of twelve members and five alternates appointed as follows:
• Five city council members and one alternate from the City of San
Jose.
• Three city council members and one alternate selected from among
the cities of Los Altos, Mountain View, Palo Alto, Santa Clara and
Sunnyvale and the town of Los Altos Hills.
• One city council member and one alternate selected from among
the cities of Campbell, Cupertino, Monte Sereno, and Saratoga and
the town of Los Gatos.
• One city council member and one alternate selected from among
the cities of Gilroy, Milpitas, and Morgan Hill.
• Two members and one alternate from the county board of
supervisors.
In an area where high technology (high-tech) industries dominate the
economic landscape, leaders of industry may be expected to be at least
informally influential in shaping transportation policy. Additionally,
environmental groups often seek to influence transportation policies.
Although voter concern and dissatisfaction with the then current state of
transportation in Santa Clara County was widespread at the time of the 1996
proposal, a lack of clear consensus existed about what, if anything, ought to be
done about the problems facing the area. This is perhaps best illustrated by the
fact that public opinion polls taken several months before the ballot indicated
that the tax increase measure was supported by only about 30% of the
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
57
electorate (San Jose Mercury News November 11, 1996). Passing a transitrelated tax was to be an uphill struggle.
Developing a transportation coalition
Leadership in support of a new transportation initiative came from two major
sources: the VTA Board, which included members of the county board of
supervisors, and the Silicon Valley Manufacturers Group (SVMG). Led by the
SVMG, supporters were recruited to include 300 organizations and key
individuals.
The significance of the support and leadership offered by the SVMG cannot be
overemphasized. Founded by high-tech pioneer David Packard in 1978, with
the support of 30 high-tech Chief Executive Officers (CEOs), the SVMG
represents 130 Silicon Valley employers, almost exclusively in the private
sector, as well as 250,000 jobs, or one-third of employment in the Silicon
Valley.
As in many areas, consensus about the need to “do something” about traffic
and congestion was not matched with an agreement about what specifically
ought to be done. The coalition of elected officials and industry adopted the
following approach toward developing a consensus for action.
(1) “95-5.” The coalition believed that consensus did exist among the
community on most major issues (the “95%”) but that past political
dialogue had instead focused on the remaining details where
disagreement persisted (the “5%”). The coalition leadership strove
to focus discussion on those areas where there was broad
agreement.
(2) “Build ripples.” Coalition leaders consciously attempted to enlarge
the movement in support of the measure “from the center out.”
(3) “Replace dogma with data.” The leadership agreed on the strategy
of relying heavily on survey and other market analysis data. This
included allowing potential opponents the opportunity to help
design survey instruments, as described below.
Due to the efforts of the initiative supporters, opposition to the 1996 measure
was sparse and isolated and an opposition coalition did not emerge.
Supporters banded together under the group identity of “Citizens Coalition for
Traffic Relief” but the contribution and leadership of the SVMG were
paramount.
Creating a transportation initiative
According to interviews with former members of the Board of Supervisors, the
selection of projects was largely predetermined from the status of, and progress
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
58
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
on, prior projects, going back to the development of the first master plan
developed in 1974. For the most part, therefore, the initiative was not designed
to fund radically new transportation modes or routes, instead, it tended to build
on existing plans, facilities, and routes.
Perhaps of greatest importance was the nature of the last measure approved by
voters, the 1992 Measure A, which focused on transit. The main reason for this
was that there were significant improvements made in the county highway
system during the late 1980s as a result of a measure passed in 1984 that
focused exclusively on highways. Thus, there was little reason or demand to
make significant highway improvements in 1992. At the same time, many
believed that funds were needed to move forward on transit, particularly light
rail along corridors identified in previous master plans.
Surveys were conducted and focus groups formed to determine the
characteristics of a successful initiative. Interestingly, the supporters of the
initiative actively invited potential opponents to collaborate on the design of
the various survey instruments. The effect of this was to allow those perceived
as “extremists” to have their ideas tested (and ultimately rejected) in polls.
Supporters of the measure believe that this eventually co-opted much of their
potential opposition, as well as serving to help make the surveys more
convincing.
One major consequence of relying on surveys to shape the design of the
proposal was that the amount of funding in the 1996 measure was significantly
less than the level requested in 1992, primarily due to the length of time
involved: the 1996 measure was for nine years and the 1992 measure was for
20 years. Consultants who helped research public for the initiative indicated
that survey results determined the maximum possible time frame and that 10
years was a threshold at which voting support diminished significantly.
Given that the funding level was to be significantly lower in 1996, the key
issue was to determine which projects involved in the 1992 measure should be
cut from the 1996 measure. The proposal could thus be viewed as an
incremental adjustment to the older initiative and not an entirely new proposal.
Because almost all the funds in the failed 1992 measure went to transit, it made
more sense to cut funds from transit. Essentially all of the highway projects
from 1992 carried over to the 1996 proposal. In fact, a few additions were
made for highways. Survey results suggested that highway improvements
would make the proposal more popular with voters. Such low cost projects as
filling potholes and signal synchronization proved to be popular with survey
respondents, and could be viewed as “pot sweeteners” for the overall package.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
59
For a variety of reasons, the 1996 proposal did not generate much
disagreement among potential opponents, particularly those who would
advocate exclusive emphasis on highway construction. One important reason
is the historic sequence of transportation initiatives. The first time that the
public was asked to vote for funding for transit in the county was 1976.
Because highway funding was provided separately by the state and federal
government, it is possible that highway advocates did not see the local funding
source for transit as a direct threat, particularly since highway improvement
projects had been approved for funding at the federal and state level.
The next election involving voter approval of funding for transportation was in
1984. Two significant developments occurred between 1976 and 1984. First,
federal and state funding for approved highway projects diminished
significantly beginning in 1980 during the Reagan administration. Second, the
county was able to obtain federal funds for transit in 1982. As a result of these
two events, with the exception of some environmentalists, transit advocates
realized that a strong case could be made for additional local funding for
highways. Thus, transit advocates did not stand in the way when a coalition
consisting of the Chamber of Commerce, the SVMG, civil engineers, and a key
member of the board of supervisors placed a sales tax measure on the ballot
that provided funding only for highway projects.
The above two elections served to establish a pattern of alternating between
measures that supported transit and measures that supported highways. This
pattern continued in 1992 where the vast majority of the requested revenues
were earmarked for transit. However, once the 1992 measure failed to obtain
the super majority that the courts determined was needed, the pattern of
alternating between transit and highways was broken. This pattern may have
served to push designers of the 1996 proposition toward an approximately
equal balance between transit and highway funding.
An important strategic decision was to spread proposed construction of light
rail across the county to three existing lines, rather than focusing funds on
completion of a single line. Survey results suggested that such an approach
increased support for the proposal. Additionally, by blending highway
spending into the proposal, a more distributive package was created. Voters
from each part of the county could therefore identify specific provisions that
would benefit them, combating the perception that funds were focused on
centralized light rail enhancements.
Overview of transportation initiative
The resulting proposal included the following provisions:
• construction of sections of three additional light rail lines
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
60
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
• widening of several key freeways, including additional HOV lanes
• synchronization of expressways
• increased commuter rail service (on existing tracks)
The preceding items were described in the advisory-only Measure A. The
actual tax increase was included in the companion Measure B, which called
for:
• a nine year, one-half cent general sales tax increase, and
• revenues over the nine year period of approximately $1 billion
(approximately $363 million to be paid by businesses and $737
million by consumers)
Additionally, Measure B called for creation of an independently appointed
“Citizens’ Watchdog Committee” to ensure that funds collected by the sales
tax increase were properly spent, along with published audits of the tax
expenditures.
Devising and implementing a marketing and campaign strategy
The overall strategy was to “work towards consensus.” This was to be
achieved by giving certain powers to members of the coalition, so that they
could bring in their constituencies. The following efforts were made to elicit a
broad based coalition of support.
1) Finance Committee (Co-chaired by a county supervisor and an executive
from a major computer hardware manufacturer)
Each member of the committee was responsible for raising a certain amount of
money. This was a critical aspect of the effort to “empower” each participant.
Presumably, individuals who had helped to raise funds would be more ardent
supporters who would help elicit support from their respective constituencies.
Ultimately the group raised more than $1.1 million, 80 percent more than had
been raised for the failed 1992 measure. Although many contributions came
from individuals, environmental groups such as the Greenbelt Alliance and the
League of Conservation Voters also made donations. Funds from large, hightech companies, which comprised nearly 90 percent of the donations and 18 of
the top 20 donors, were critical to the success of the fundraising effort.
2) Speaker’s Bureau
The coalition created a speakers’ bureau that could personally deliver the
message about the ballot initiative. According to coalition leaders, coalition
members made 200 appearances, to audiences totaling 10,000. The speaker
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
61
was “marketed” to fit the audience such that an appropriate speaker was
identified for each group.
3) Editorial Board
The editorial board wrote editorials for 35 newspapers and received
endorsements from 34. Again, an effort was made to “market” the writer to the
individual audience served by each paper. Editorial authors were matched by
backgrounds to the publications selected for the campaign.
4) Marketing within companies.
Campaign supporters directly addressed 103 companies who represented
250,000 employees. A previous internal lobbying effort by the SVMG to
combat frivolous lawsuits against high-tech companies had recently received a
lot more publicity, but according to coalition leaders the effort to pass Measure
B was much bigger and more intense. The campaign consisted of nine
“loaned” executives who appealed to their corporate colleagues through email,
voicemail, corporate newsletters, and payroll mailings, each appropriate to the
individual corporate culture. They also conducted in-house employee
education campaigns.
5)
Mass marketing
In addition to “free media” (e.g., press conferences), the campaign relied
heavily on, direct mail, television, and radio ads. More details about this aspect
of the campaign follow.
The general strategy was to target “yes” and swing voters, and to avoid “no”
voters (so as not to “energize” them). The message was designed to tease out
the existing motivation to support the tax. This was achieved by identifying
what voters wanted and making sure they heard a coordinated message. The
emphasis was to be the “carrot” (improvements) and not the “stick” (higher
taxes).
The campaign was directed by consultants from the firm of Townsend,
Raimundo, Bessler, and Usher. Another firm conducted a series of surveys and
focus groups that were used in ways discussed later in this section. Great care
was devoted to both the wording of the paired measures and to the wording of
measure descriptions in promotion and advertising.
There were three general key tactics.
1) The sequence of projects described in Measure A. The ballot language was
constructed so as to place popular projects in the proper sequence in the
description of Measure A. The research suggested that the most important ones
were the first, second, and last ones indicated. Thus first item mentioned was
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
62
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
fixing streets and potholes. The second one was a link to the regional BART
transit system, and the last one was improved transit for seniors and disabled
people. One of these appealed to drivers, one to transit users and supporters,
and the other for “disadvantaged” groups that tend to be viewed
sympathetically.
2) An emphasis on accountability in Measure B. The second key tactic was to
emphasize accountability in Measure B, since it was a general tax measure.
This was achieved in a number of ways. The term “mandatory restrictions”
was used to describe spending under Measure B. To assure accountability, a
“citizens watchdog committee”, selected independently, would conduct yearly
audits and report the results to the public.
The specific expiration date was also intended to reassure voters. Research
discovered that the expiration date needed to be fewer than ten years but that
there was little sensitivity between a nine year expiration date and shorter
expiration dates. The surveys and focus groups played an important role in
determining the description of the measures and the transportation projects
listed.
3) Avoid mention of “taxes.” The third key tactic involved avoiding the term
“tax” unless it was absolutely necessary to use the term. The basic concept
involved stating that both measures A and B needed to be passed “in order to
get traffic relief.” Descriptions of Measure B indicated that it provided the
funding for the transportation projects.
The use of selective targeting was combined with a selective message. This
selective targeting was implemented through the heavy use of direct mail.
Voters likely to support the measure were targeted to encourage them to
actually vote. The second group targeted was swing voters. Mailers were not
sent to those likely to vote against the paired measures. (This was part of the
“run silent, run deep” philosophy discussed in more detail under the next
success factor.) The underlying reason for this philosophy was to avoid stirring
up the opposition. The characteristics and identities of these two groups of
targeted voters were determined through the surveys conducted.
Direct mail was emphasized because it reaches a selective target with a
selective message. Six different mailings were used. The first one, which was
sent approximately one month prior to the election was a general message.
The second one was targeted to five separate age categories and had agespecific photos in each mailer. The third and fourth mailers were sent to 30-35
geographic segments. They included local projects and were from a “neighbor”
supporting the measure. The fifth mailer emphasized the endorsement of the
Mercury News along with other supporting organizations. The surveys were
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
63
used to help determine which organizations to include in the mailing. The sixth
mailer provided a summary.
Television advertising also was used, although the campaign consultants did
not believe it would be effective. In fact, use of television arguably ran
somewhat counter to the “run silent, run deep” strategy adopted by the
campaign. However, the television advertising used was very general and did
not mention the tax increase, but instead pointed out that the measure would
fund the repair of potholes and reduce congestion.
Market research efforts were central to the supporter’s campaign. Research
was first used to determine the wording and positioning of the measures.
Then, during the campaign itself, survey research helped to determine the
wording and positioning of the mailers, including which endorsements to
emphasize, and the sequencing of the endorsements listed. Finally, research
helped to identify, along with the precinct data, which positive and swing
voters to target with the mailings.
Voters were asked how they would vote under both different wordings of the
measures as well as under different arguments for and against the measures.
Respondents who continually indicated they would vote for passage were
classified as positive voters and those people who varied their vote depending
on the specific wording used and the arguments presented were classified as
swing voters. Correlation analysis was used to determine links between voting
tendencies and a variety of personal characteristics, in order to determine what
personal characteristics were most highly correlated with voter tendencies.
Finally, voter-specific personal characteristics were linked to precinct lists to
identify the specific households to send the mailers to.
Opposition efforts
Due to the pointed efforts of the coalition in support of the measure described
earlier, organized opposition was scanty. Some local environmentalists
supported the measure, some remained on the sidelines. Opposition was
largely confined to taxpayer groups, who routinely oppose nearly every local
tax proposal but typically lack sufficient funding and organizational resources
to mount effective campaigns.
Media coverage concerning the election was limited, presumably due to
interest in the various concurrent state and national elections. The local
newspaper did endorse the twin measures, although it did print opposing
views. For the most part, however, the coverage of measure was neutral and
even favorable and included articles that documented (accurately enough)
business support for the measure (see, e.g., San Jose Mercury News October
13, 1996). Because of this limited and primarily favorable coverage, supporters
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
64
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
of the tax measure were operating largely in a vacuum where their viewpoint
would not receive a serious challenge.
Outcome evaluation
On November 5, 1996, voters in Santa Clara County approved of both
measures A and B. Whereas the advisory measure A, which described the uses
of the new funding, passed 76%, the actual tax measure B passed by only 52%.
By contrast, Measure A in 1992 had passed by 54% amid the economic
backdrop of a serious recession. Exit polling revealed that, the nonpartisan
character of the initiative process notwithstanding, Democrats voted for the tax
increase 63% to 37%, while Republicans voted against it 57% to 43% (Gerston
1977). The exit poll also revealed that wealthier individuals tended to favor the
measure, and vice versa.
Thus, although the passage of Measures A and B was a nominal success for its
supporters, the small margin of victory, achieved in a context wherein nearly
every possible intervening circumstance was favorable, is equally notable.
Indeed, the Santa Clara County case illustrates the great difficulty advocates of
public transportation tax initiatives face, particularly in the instance of a
general sales tax. Consider that proponents in this instance were working with
the following advantageous circumstances:
• Lack of well-organized opposition
• Historical support for transportation initiatives and a liberal-minded
electorate
• Local economic boom
• A general election environment
• Support and leadership from the local business community, included a
million dollar campaign fund
Nevertheless, careful consideration of the Santa Clara case suggests that the
effectiveness of certain strategies that may be applicable to other communities,
including the following:
1) Co-option of potential opposition. The public transportation supporters were
able to achieve this in several ways. First, they invited potential foes to
participate in the market research that helped shape the final ballot measure.
Once convinced that their own proposals had no chance with the public,
opponents were more likely to stand aside or even support the proponent’s
approach. The mixture of highway and transit spending placed on the ballot
helped minimize opposition from ardent highway supporters.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
65
2) Use of targeted direct mail. The proponents campaign consultants used
survey research data to fine tune their message to various demographic groups.
Speakers and guest editorials were also targeted to match the messenger to the
audience.
3) In-house direct marketing. Supporters marketed the initiative to their own
employees in major companies in the Silicon Valley, a strategy that gave the
measures great credibility in a key demographic segment.
4) “Run silent, run deep.” Initiative opponents eschewed direct mass media
publicity for the measure, focusing their efforts on direct mail and in-person
appeals to potential voters. This approach lessens the possibility that voters
will be exposed to potentially dissuasive information.
Although the effectiveness of these strategies cannot be verified, it is plausible
to expect that without each of them the Santa Clara County tax measure might
well have failed, despite the many favorable circumstances it faced.
CASE STUDY II
Sonoma County: Defeat of funding for highway widening and rail (1998)
For reasons discussed previously in the Santa Clara County case, Sonoma
County used the paired measure approach in attempting to obtain voter
approval for transportation improvements. One of the measures involved a half
cent sales tax for general county purposes with a 20-year expiration date. The
paired advisory measure identified the recommended transportation projects.
The major projects involved widening Highway 101, the one major corridor in
the county and implementation of a passenger rail service running along the
same corridor. The advisory transportation measure passed by a vote of
104,129 (72.4%) to 39,763. However, the sales tax measure failed with 68,062
voting in favor of the tax (47.6%) and 75,051 voting against it.
This case has some interesting lessons pertaining to a) coalition building, b)
selection of a transportation package and funding mechanisms, c) marketing
strategies and tactics, and d) the impact of a state political structure dealing
with funding and taxes.
Interviews with key participants and documents provided the information for
this case. Interviews were conducted with members of the citizens’ coalition of
business people and environmentalists that developed the transportation
package presented to the county board of supervisors, supporters who worked
to pass the paired measures, and opponents who worked to defeat the
measures. In addition, the political consultant interviewed specifically for the
Santa Clara County case also represented the proponents in Sonoma County.
Key documents used include a) the final report of a Sonoma and Marin County
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
66
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
Multi-Modal Transportation and Land Use Study, b) copies of direct mail used
by both sides, c) newspaper articles and editorials in the newspaper with the
largest circulation in the County, d) results from post-election questionnaire,
and e) a breakdown of expenditures by proponents of the paired measures.
Background: geography and population profile
Sonoma County is immediately north of Marin County and which is
immediately north of the Golden Gate Bridge. With a population of 443,669,
the number of people in the county is significantly less than the number in the
other three case areas studied. Santa Rosa is the largest community in the
county with a population of 138,742 and has increased by approximately 100
percent since 1970.
Geographically, Sonoma County is quite large and rather sparsely populated.
The main corridor, Highway 101, runs approximately 50 miles from the
northern county border to the southern county border and about 50 miles from
the ocean to rural areas of the wine country. Approximately 70 percent of the
population live very close to the 101 corridor (DuBay August 2, 1999).
Interviewees described the population as not having a strong anti-tax
philosophy and also having a strong environmental and anti-growth
perspective. This assessment of the population is supported by both a survey
conducted about one month after the election and by a vote in 1990. In 1990,
the voters approved a 20-year increase in the sales tax of one quarter of a cent
to fund open spaces protection and enhancement in the county. This result
demonstrates a willingness both to pass a significant tax and to protect the
environment and control growth. In the survey conducted after the election,
62% of the 400 respondents responded “yes” to a question asking if they
considered themselves to be an “environmentalist” and 72% considered
themselves to be either no-growth (39%) or slow-growth (33%). At the same
time, the population does not seem to exhibit a high level of trust in
government. Although the post-election survey did not directly ask
respondents to indicate their degree of trust in government, it did ask people to
identify whether each of 17 reasons presented were major, moderate, or minor
reasons for the sales tax failing. The reason with the highest percentage of
responses in the “major” category was “people didn’t trust government’s
ability to follow through on its promises” (63%).
Background: transportation and traffic
Highway 101 corridor is the major corridor in the county. This is a separated
highway with two lanes in each direction. No HOV lanes currently exist on
101 within Sonoma County but they do exist in Marin County. In 1995, two
segments of 101, approximately 10 miles each, were classified as having an
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
67
“F” level of service during commute hours. This is an overcapacity
classification with stop-and-go traffic (Calthorpe 1997). By comparison,
virtually all of 101 in Marin County is classified at a “F” service level during
commute hours. Due to the number of popular tourist and recreational
destinations within the county, 101 is usually congested on the weekends
(Calthorpe 1997).
Bus transit is available in the county. The mode split for transit currently is two
percent, which is quite low (Calthorpe 1997). Work commuting is mostly
within the county: Eighty-two percent of employed county residents work in
the county, compared to fifty-nine percent for Marin County (Calthorpe 1997).
Given that the vast majority of residents work within the county, commutes
tend to be short to moderate in length. Although many of the jobs, as well as
most of the residents, are close to the 101 corridor, job locations are not highly
clustered (Calthorpe 1997).
The 101 corridor transportation system in place prior to the election was
similar to the system that existed during the 1970s and the 1980s, when the
population was about 220,000 less than it was at the time of the 1998 election.
Background: relevant historical events
In 1990 an attempt had been made to obtain voter approval for a one-quarter of
a cent increase in the sales tax for transportation improvements. The money
would have been used primarily for capacity improvements on 101. The
business community actively supported this measure and perceived that they
needed the environmental community to at least stay neutral on the
transportation tax. They worked with the environmental community to try to
achieve this goal by having a second proposition submitted for voter approval
that the environmental community strongly supported. This second proposition
involved a one-quarter of a cent sales tax to fund open spaces. The business
community thought that the environmentalists would stay neutral on the
transportation measure but many environmentalists actively opposed the
transportation measure and it lost by a 54 percent to 46 percent vote.
The primary conclusion developed from this failed attempt appears to have
been that success would be likely in the future if a transportation package
acceptable to both the business and environmental community could be
developed. At the same time, interviewees indicated that the 1990 election
made it very difficult for the business and environmental community to work
together to obtain local funding for transportation for a number of years.
In 1995, the California Transportation Commission approved more than $60
million to purchase the Northwestern Pacific Railroad (NWPRR) right-of-way
and provide infrastructure improvements along the railroad tracks. This event
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
68
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
was significant because it created the opportunity for a mixed transportation
package including both rail transit and highway improvements. Thus,
according to three people interviewed, people representing both the business
and environmental communities began to work together to develop a second
transportation package.
Also in the mid-1990s, Caltrans, the Sonoma County Transportation Authority,
and the Marin Countywide Planning Agency decided to sponsor a two county
(Sonoma and Marin) transportation and land use study along the 101 corridor.
The final report of this study, completed by Calthorpe Associates, was
published in 1997. It identified four transportation improvement alternatives
and conducted a cost/benefit analysis of these alternatives. The recommended
package included implementing a rail transit system with feeder bus service
and widening Highway 101 by one lane in each direction in some segments
within the county. The extra lane would be used as a HOV lane during rush
hours. This recommendation had elements appealing to both business and
environmental interests. The Sonoma County components of the package
recommended in this report were used as a framework for the citizen task force
working on a transportation package and funding plan. The Calthorpe report
also was helpful in providing cost and use projections with both a baseline
system and each alternative scenario.
The successful passage of Measures A and B in Santa Clara County,
previously discussed in detail had significant impact on the 98 measures in
Sonoma County. The success in Santa Clara County created a sense of
optimism for transportation proponents in Sonoma County. In addition, the
Santa Clara framework and marketing approach were used as a model for
Sonoma County. One component of the Santa Clara model was the
development of a transportation package by a citizens’ coalition. Members of
the Sonoma County citizens’ coalition met with Carl Guardino who helped
lead the coalition in Santa Clara. The Sonoma Coalition concluded that they
should use this approach.
In conclusion, although the 1990 election created a temporary split in a
business and environmental coalition, events in 1995 and 1996 gave this
coalition both the motivation and the means to work together to make a second
attempt at obtaining voter approval to fund transportation improvements. This
coalition, which called itself the Citizens for Traffic Relief (CTR) decided to
develop a transportation package and funding mechanism that would be
presented to the county board of supervisors for approval. If approved, the
package, with possible modifications from the Board, would then be presented
to the voters for approval.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
69
Development of the transportation package by the CTR Coalition
Once the Calthorpe study was completed in mid-1997, the CTR coalition was
able to reach closure quickly and easily. This was because this study had
components that were viewed positively by both groups. In addition, according
to a co-chair representing the business community, business representatives
seemed to believe in smart and controlled growth and were supportive of
transit. Coalition representatives from both segments also were pragmatists
committed to developing a package.
In addition, the CTR coalition commissioned a survey in late 1997 to test
reaction to the Calthorpe study’s recommended package. The study results
reinforced the prevailing view that both a highway and a rail transit component
were needed for voter approval. According to a co-chair interviewed,
respondents’ approval ratings dropped below 50 percent if either the highway
widening or the rail transit components were omitted.
During the course of their meetings, the CTR coalition became aware that one
group, the County Taxpayers Association, would not support the package
(Ellman August 2, 1999).1 They were in favor of widening 101 but were
opposed to HOV lanes and to the rail transit line. They also were in favor of
congestion pricing of some sort and thought that much of the financing should
be through the gas taxes to the state and federal governments. A leader of this
group indicated that they probably would not have opposed the measure if
transit had been excluded and if the new lanes on 101 had not been restricted to
HOV use during commute times. Of course, the environmental representatives
in the coalition would not support this position and the taxpayers association
did not appear willing to compromise.
The measure presented to the county board of supervisors consisted of the
following components:
• Widening of 101 by one lane in each direction through most of the
County but excluding both a segment in the north and one in the
south. The extra lanes would be dedicated to HOV use during
commute hours.
• Improving or constructing 12 interchanges along 101.
• Providing safety improvements to Highway 116, both west and east
of 101.
1
Two other groups did oppose the measures but the CTR coalition had little reason to be aware
of their opposition during their deliberations (Ellman August 2, 1999). One group was the
Environmental Defense Fund (EDF) and the second group consisted of some local
environmentalists who did not seem to be leaders in any of the formal environmental groups.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
70
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
• Implementing rail service on the NWPRR line by upgrading the
tracks, constructing stations, purchasing commute trains, and
providing operating expenses. Both freight and passenger service
would be provided.
• Expanded bus service with emphasis on feeder lines with the rail
line.
• Building and improving bike and pedestrian paths.
The above package was quite similar to the package recommended in the
Calthorpe report. However, Calthorpe recommended less extensive use of
HOV lanes. The report also stated that a “bus/HOV lane is the least cost
effective transit investment” (Calthorpe 1997 p.128) and later stated on the
same page that “the benefits of HOV lanes vary along the corridor.”
Unanimous approval of a modified package by the County Board of
Supervisors
The County Board of Supervisors has five members who are elected through
geographic district elections. According to a proponent interviewed, the Board
consisted of one environmentalist, one conservative person from the southern
part of the County, and three moderates during the relevant time period.
The Board added some things to the package that increased the projected costs
by approximately $183 million to a total of $948 million (Sonoma County
Citizens for Balanced Transportation “Fact-Pack,” 1998). More specifically, it
added widening of 101 from Petaluma to the southern county line (a projected
cost of $89 million) along with storm repair, safety, and street maintenance
improvements in parts of the County both east and west of the 101 corridor (a
projected cost of $94 million) (Ellman August 2, 1999). These changes
allowed each board member to indicate that the package included something
for voters in their district.
The most controversial modification was the widening of 101 south of
Petaluma. This issue was discussed in a series of articles in the Santa Rosa
Press Democrat in the first quarter of 1998.2 Even environmentalists in the
CTR coalition were quoted as being opposed to this widening believing it
would result in population growth. Many voters were strongly in favor of this
widening. The board developed a compromise that added the highway
widening but also added a county measure to the November ballot that
2
The Santa Rosa Press Democrat is the main newspaper in the county and it had an important
role in the campaign that will be discussed in more detail later in the chapter. The abbreviated
name of the Press Democrat will be used for this newspaper.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
71
restricted population growth south of Petaluma (Wilford July 23, 1999). The
board voted unanimously to approve the modified package in May of 1998.
Cost projections for each of the major components of the transportation
package approved by the Board are identified below (Sonoma County Citizens
for Balanced Transportation “Fact-Pack” 1998).
•
•
•
•
Highway 101 widening - $382 million
Highway 101 interchanges - $186 million
Highway 116, streets and roads - $130 million
Rail, bus and bike - $250 million
The campaign strategy used by proponents of Measures B and C
The proponents used the political consultants who were used by Santa Clara
County in their successful attempt to fund a combination of highway and
transit improvements using the paired measure approach. The consultants
essentially used the same strategy and tactics in Sonoma County as they used
in Santa Clara County.
According to the political consultant, the basic strategy involved running a
stealth campaign that tried to prevent vocal opposition from forming. This
strategy was based largely on the result of a survey done months prior to the
election. In this survey, a significant majority indicated they would vote for
both measures (DuBay August 2, 1999). If organized vocal opposition could
be minimized, voter intentions would remain stable through the election as
they did in Santa Clara County (Bessler December 9, 1998). This was a
reasonable initial strategy. However, since vocal and effective opposition
cannot be prevented, developing a contingency response plan if this opposition
developed would have been prudent. Proponents indicated that a
comprehensive aggressive contingency response plan did not exist.
Direct mail was the primary component of the campaign for two reasons. First,
since it was a less public medium than television, radio, or newspaper, it was
more consistent with the stealth concept. The sales tax increase would be
reduced from .004 to .002 once the bonds were retired. Different bond
retirement dates were identified, with the earliest date being 2020 and the latest
being 2035 (Denver Post October 17, 1997).
The direct mail campaign in Sonoma County was very similar to that used in
Santa Clara County. Since this campaign already has been described in the
Santa Clara case, it will be only briefly described here. Five mailers were
involved with the first one sent about one month prior to the election. They
were sent to 95,000 households representing about 130,000 registered voters
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
72
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
who were likely to vote in the November, 1998, election, based upon past
voting behavior. Except for the last mailer, which was a postcard, the mailers
had a great deal of information on them. The main points were that the paired
measures were supported by both environmentalists and the business
community, and that voting for both measures was needed and would provide
traffic relief. In addition to making these points directly, they were supported
by a) rather long quotes from supporters in the business and the environmental
communities, b) a list of organizations from both camps that endorsed the
paired measures, c) a somewhat complicated map of where improvements
would be made with a list of projects, starting with the highway 101
improvements and passenger rail service.
One component of the stealth campaign involved trying to keep the issue out of
the news (Bessler December 9, 1999). Thus, according to proponents
interviewed, they made little, if any, effort to obtain news coverage in either
the Press Democrat or the broadcast media.
The campaign strategy used by opponents of Measures B and C
Three groups of opponents worked together to defeat the paired measures. One
group was the Sonoma County Taxpayers Association. Although proponents
had some reason to hope that this group would not actively oppose the
measure, since over two-thirds of the money involved highway projects, which
this association supported, their opposition was not a surprise. A second
informal group consisted of local environmentalists. Their active involvement
started in August, according to one of the members. Given the heavy emphasis
on highway capacity increases, particularly south of Petaluma, which
environmentalists had actively opposed, it should not have been surprising that
some individual environmentalists opposed the paired measures. The third
group opposing the measures probably was both the most critical and
surprising group.3 This was the Environmental Defense Fund (EDF). Their
involvement was surprising both because they had little prior involvement in
Sonoma County and because they had become actively involved in opposing
only one prior campaign, and that one, Alameda County, was in early 1998.
The basis for EDF’s involvement is described below.
EDF believes in user fees for automobiles, such as congestion pricing. They
tried without success to persuade different communities to implement
congestion pricing. They then decided to use a new tactic to accomplish their
objective, namely active opposition to voter packages that did not include
congestion pricing. The June, 1998 election in Alameda was their first effort at
3
Although they did not oppose the paired measures, two major environmental organizations,
the Sierra Club and the Greenbelt Alliance, remained neutral in the campaign.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
73
such opposition. Their only involvement in Sonoma County prior to the 1998
election was a study of congestion pricing on 101 conducted by the
Metropolitan Transportation Commission. (This commission is involved in
transportation in nine counties in the larger Bay Area, including Sonoma
County.) The study apparently concluded that using the new lanes on 101 for
free HOV use, combined with paid use by solo drivers, was “financially,
physically, and operationally feasible.” (Press Democrat July 9, 1999)
EDF probably would not have opposed the paired measures in Sonoma County
if congestion pricing had been used to help finance the transportation
improvements. The primary reasons stated by proponents interviewed for not
including congestion pricing to help fund the transportation improvements
were that it would not have raised sufficient revenues by itself to fund the
improvements and that voters were opposed to the concept. Both of these
arguments can be questioned. The first argument does not explain why
congestion pricing was not part of the funding package. The second argument
possibly could have been countered since it would have been less restrictive
than what was included in the paired measures. In other words, giving solo
drivers the option of using HOV lanes at a known price, rather than not
allowing them to use the lane under any conditions, seems to be something that
voters might not oppose if it were appropriately positioned. At the same time,
although the congestion pricing study seemingly began in 1997 and proponents
certainly were aware of the study, it may have been too late to incorporate this
recommendation of the study into the paired measures, given that the study
was not published until after the county board had already approved the
measures presented to the voters.
Since congestion pricing was not included in the paired measures, EDF
decided to become involved in the opposition to the paired measures and their
involvement started sometime during August and September. All opponents
interviewed, as well as proponents, indicated that the EDF played a critical role
in the defeat of the paired measures. More specifically, all agreed that they
organized the opponents through weekly meetings. In addition, their research
skills, combined with their name, helped the opposition obtain some prominent
coverage in the Press Democrat during the last few weeks prior to the election.
Since the opposition raised only about $15,000, they had to rely on unpaid
methods to communicate their opposition. The only exception to the unpaid
means involved one direct mailer to about 50,000 households (Hall and Taylor
August 2, 1999). The unpaid means involved concerted efforts to have the
Press Democrat publish articles in the news section that presented their
arguments, grass roots efforts such as debates and visits to organizations in the
county, letters to the editor, and ballot arguments.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
74
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
The opposition used a shotgun approach in opposing the paired measures. In
other words, they presented a number of different reasons for voting against
the measures. This approach made it somewhat difficult for the proponents to
respond and also may have appealed to a larger segment of voters. The
arguments used by opponents are presented below.
• The paired measure approach involved having a transportation
projects measure that was merely advisory since the money raised
by the sales tax went into the general fund. This was the main
message in the direct mail campaign with the headline that
“Measure C is a $650 million blank check.” A check for
$650,000,000 with a notation that the revenue could be used for
whatever the board of supervisors wanted was under the headline.
As discussed in more detail in the next section, this argument was
reinforced frequently in articles in the Press Democrat.
• The costs of the transportation projects would have been
significantly higher than projected since debt financing would have
been needed and it was not included in estimates. This was the
primary focus of a press conference held by EDF at which a
position paper was presented. This topic was the focus of an article
in the Press Democrat shortly after the press conference. It also was
suggested that, as a result of insufficient revenues, the passenger
rail project would not be finished.
• The passenger rail line project was incompletely developed and
would have no more than a minimal impact on 101 congestion due
to low ridership. This was the primary focus of a press conference
held by EDF at which a second position paper was presented with
ridership estimates presented, based upon population and
occupation destinations. The projections were lower than those in
the Calthorpe report. This “waste” argument was reinforced by the
name used by the opponents. They called themselves “Citizens
Against Wasting Millions.”
• The 101 widening would increase the rate of population growth
more than it would reduce congestion on 101. This argument was
made in a guest editorial by Laura Hall that appeared in the Press
Democrat in August. It also was made in a ballot argument signed
by Laura Hall and four other local people affiliated either with an
environmental organization, or an institution of higher learning.
This same argument also discussed the need for a “viable transit
program that gets the priority it deserves and operates frequently
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
75
enough to reduce traffic; and express carpool lanes paid for by road
users, not HOV lanes paid for by sales taxes.” It is interesting that
this ballot argument also spoke negatively of the passenger rail
transit component as the “slow train to nowhere.”
• A sales tax is a poor and unfair way to finance transportation
projects. This was a ballot argument prepared by the taxpayers
association. This argument also talked about using the new lanes on
101 as free for HOV use but toll lanes for solo drivers. It also
reinforced the message that the funds would go into the general
fund.
Newspaper coverage
Based on a review of their web site, the Press Democrat had extensive
coverage of the transportation issue in the news section of the paper while the
package was being developed in the latter half of 1997, in the first quarter of
1998, and in the last two months prior to the actual election. For example, 13
articles written from the middle of September, 1998, through the end of
October were identified. Additional articles probably were written that
research did not find.
The articles covered a number of specific topics and were balanced. However,
virtually every article researched mentioned two characteristics that reinforced
themes used by opponents. First, almost every article indicated that the sales
tax went into the general fund and could legally be used for purposes other
than the transportation projects listed in the companion measure. Second,
many articles mentioned that the paired measure approach was used to get
around state law on special taxes requiring a two-thirds vote.
The focus of one article written within a week of the election (Press Democrat
October 31, 1998) was intriguing and certainly could have helped opponents
more than proponents. This article indicated that Sonoma County would
receive $594 million from gas taxes over the next 20 years. Of course, this
article also included quotes from opponents indicating that the sales tax was
unnecessary with this significant infusion of gas tax revenue.
Geographical distribution of voting
The issue of whether transportation packages involving a significant tax need
to provide benefits to geographical segments roughly proportional to tax
revenues in each segment has been raised in prior cases. Given that most of the
sales tax revenue raised in Sonoma County would have gone to the 101
corridor, including the passenger rail line, an analysis of the geographical
distribution of votes is relevant here. This distribution indicates that voting
tendencies definitely were correlated with the distance of the voter from the
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
76
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
101 corridor. In the communities along the 101 corridor where the widening
was proposed, 50.5% of the 80,740 votes supported the sales tax measure. In
the communities not along the 101 corridor, in unincorporated areas, which
generally are not along the corridor, and in a couple of small communities on
the 101 corridor but not where the widening was proposed, 43.7% of the
62,293 votes supported the sales tax measure. Thus, the results in Sonoma
County are consistent with the interpretation that voter tendencies are based
partly on the extent that their geographical area gets benefits roughly
proportional to the taxes paid by the area.
One person interviewed compared the voting results in areas not on the 101
corridor in 1998 to those in 1990. He indicated that the results were virtually
identical even though no benefits were provided to these regions in 1990 while
about $130 million was supposed to go to these regions in 1998. Although this
comparison does not begin to approximate a controlled experiment, the results
are consistent with the hypothesis that providing less than proportional benefits
to a geographic segment seems to have little impact on voter tendencies
compared to offering no benefits to the segment.
Reasons for voter rejection
Survey Results: A random survey was conducted for the proponents about one
month after the election. Through both open- and close-ended questions, along
with scales, this survey provides direct responses from voters about their
voting behavior. Thus, it is an excellent starting point for understanding why
people supported and rejected the paired measures. Some of the key findings
are cited below, followed by an interpretation.
• In response to an open-ended question on the transportation
package measure, some “yes” voters mentioned highway
improvements while others mentioned transit improvements.
Highway improvements, however, with an emphasis on highway
101 widening, were mentioned by significantly more proponents
(54 responses) than transit improvements (seven responses).
• In response to the same open-ended question, “no” voters tended to
have different views than “yes” ones towards highways, growth,
and transit. Here a larger number of “no” voters were opposed to
the highway widening or growth perceived to be generated by the
widening (15 responses) than opposed to the transit component
(four responses).
• In response to an open-ended question on the sales tax measure,
“no” voters’ responses essentially could be put into three
categories, excluding the most common response, which was a
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
77
vague negative statement about taxes. The most common response
was that the gas tax already imposed should be used rather than an
additional sales tax (14 responses). The next most common
response was a lack of trust that the tax would be used for
transportation (nine responses). Four respondents indicated they
were opposed to highway widening.
• The two most common sources of information checked in response
to a close-ended question were literature in the mail (70%) and
newspapers (64%). These results are consistent with the prior
discussion of heavy coverage in the primary newspaper plus use of
direct mail by both proponents and opponents.
• Respondents were presented with 17 possible reasons for the sales
tax measure failing and asked to indicate with each reason whether
it was a major, moderate, or minor reason for failure. As mentioned
previously in the background section, the reason with the largest
percentage of “major” responses was the statement “people didn’t
trust government’s ability to follow through on their promises”
(63%). The statement “transportation plans should be funded by the
gas tax or some other type of tax, not by a sales tax” had the second
highest percentage of responses in the major category (52%). Thus,
the two most important negatives with this close-ended questioning
format are consistent with the two most common reasons cited in
the open-ended question asking voters who voted against the sales
tax why they did so. Since these two reasons were the most
important ones identified in the post-election survey, each is
discussed in more detail below.
• One of the main arguments used by the opposition was that the rail
line would be a waste of money and have little impact on highway
congestion due to low ridership. Since this argument is consistent
with the perceptions of at least some segments of the population, it
could have been hypothesized that this would have been mentioned
by some “no” voters as a primary reason for their vote. However,
this reason was rarely given in the open-ended responses.
Unfortunately, none of the 17 possible reasons for the loss of the
sales tax measure involved a statement about the rail line probably
having only a small impact on highway congestion. Thus, both the
credibility and salience of this argument on voter tendencies
remains uncertain.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
78
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
The paired measure concept combined with voter distrust of politicians
Voters realized that the sales tax would go into the general fund and that the
vote on the transportation package was advisory only. As indicated in the
survey discussed above, many voters seemed to doubt that the sales tax
revenue would be used solely for transportation projects or that all the
transportation projects listed would be funded with the sales tax revenues
raised.
Given that the paired concept succeeded in Santa Clara County with the same
marketing and communications strategy as used in Sonoma County, it is
relevant to discuss why it succeeded in the former county but failed in the latter
one. Two plausible explanations exist. First, unlike Santa Clara County,
organized opposition existed in Sonoma County and this opposition focused
heavily on the “blank check” and likely cost overrun arguments in their direct
mailer, a press conference, and in ballot arguments. Second, the primary
newspaper in Sonoma County had extensive coverage of the paired measures
with frequent mention of the nonbinding nature of the paired measures. Given
the very close vote in Santa Clara County, the paired measures may not have
passed there if organized opposition or extensive newspaper coverage had
existed.
Although the nonbinding nature of the sales tax measure was the primary
reason given by voters for opposing it, some of these voters probably would
have voted in favor of the tax if they perceived no additional problems with the
measure and also saw significant benefits. In other words, many voters
probably weighed the pros against the cons in determining how to vote on the
sales tax measure. For example, some of the voters who gave their primary
reason for opposing the sales tax measure as uncertainty that the money would
be spent on the transportation projects may also have had a general opposition
to higher taxes and also perceived that rail transit is an inefficient use of public
expenditures.
Since a two-thirds vote would have been needed if the paired measure
approach had not been used, the latter was a better approach to use even though
it failed. Since the paired measure approach seems to be the best legal option
currently available in California, it is relevant to examine whether proponents
could have effectively countered this information.
The use of a “stealth” campaign by the proponents suggests they realized it
would be difficult to counter arguments such as the nonbinding nature of the
sales tax once they were raised. The one direct attempt to counter this
argument involved having the county board of supervisors take a pledge to
only spend the sales tax revenue for the transportation projects listed in the
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
79
advisory measure. If anything, this attempt probably made things worse by
calling more attention and press coverage to the nonbinding nature of the sales
tax. Even if voters believed that the board members who made the pledge
would honor their word, which is unlikely given general voter distrust of
politicians, they would have realized that these members would not be on the
board during the entire 20-year period that the tax would exist. Assuming that
it was best for proponents to avoid directly countering the nonbinding issue,
the only reasonable response seems to have been the one used in Seattle. This
involves attacking the opponents as not having a solution to a transportation
problem that was bad and would get even worse and more expensive to fix if
not fixed promptly. This attack was made in an editorial in the Press Democrat
shortly before the election, but it might have been more effective if it had been
made more repetitively through a media campaign.
The gas tax issue
Some “no” voters thought that the gas tax money they were already paying for
transportation, estimated at about $70 million per year, should be sufficient to
pay for some, if not most, of the transportation improvements. In other words,
they perceived a combination of a significant gasoline tax plus a half-cent sales
tax for 20 years to be too expensive for benefits that opponents said would be
minor and short-term. Although this gas tax issue was not directly mentioned
by opponents in their ballot arguments or their direct mailer, it was mentioned
many times in the Press Democrat, including an article within a week of the
election. In addition, opponents indirectly focused on this issue through their
arguments that neither the highway widening nor the rail transit line would
lessen congestion on 101 to any significant extent. This is a relevant indirect
response because it addresses the issue of value of the “additional” tax.
The gas tax issue is one that probably could have been addressed more
effectively by proponents in their marketing and communications campaign.
For example, no mention was made in their mailers that approximately onethird of the costs would be funded by gas tax revenues with two-thirds being
funded by the sales tax. They also could have effectively argued that they were
likely to get a greater infusion of gas tax revenue if they agreed to share the
costs through a sales tax.
The gas tax issue might have been indirectly diffused if a smaller
transportation package had existed. The gas tax revenue would have covered a
larger percentage of the funding with the sales tax covering a smaller
percentage with a less costly package. In addition, a shorter sunset package
could have been used with a smaller and less costly transportation package.
Since uncertainty tends to decrease with shorter time frames, voters might have
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
80
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
felt more confident about how the board would spend the money, as well as
more certain about costs, with a shorter sunset period. This issue will be
discussed in more detail in the concluding chapter.
Inclusion of both highway widening and transit elements
Responses to the survey discussed previously suggest an underlying dilemma
in developing a transportation package that will obtain voter approval and also
minimize organized opposition. This dilemma is that public perceptions
towards modes of transportation are somewhat polarized with some people
being anti-transit, and others being anti-highway.4 People in both extremes
voted against the sales tax because it included the component to which they
were opposed. Since most of the revenue raised would have gone to highway
improvements in Sonoma County, the anti-highway people were more likely to
vote against the sales tax measure than the anti-transit people. Organized
opposition also came from both groups. Local environmentalists who worked
to defeat the measures were strongly opposed to the highway widening. The
leader from the taxpayers association who was interviewed indicated that their
group was strongly opposed to the rail transit component.
Given the polarization identified in the previous paragraph, it seems
unreasonable to assume that opposition to any transportation package will not
occur. Thus, although a “stealth” campaign may be appropriate before
opposition develops, it is critical to have a contingency plan when and if
opposition does develop. This concept will be developed in more detail in the
concluding chapter.
The polarization of views toward modes of transportation also created a
challenge in developing a transportation package. Given the number of
supporters of highway widening, it definitely was necessary to have this
component to obtain voter approval in Sonoma County. Although very few
supporters of the measure indicated in the survey that they did so because of
the transit component and although opposition to the measures came from
environmentalists, it also probably was wise to include the transit component
in the package. This conclusion is made for several reasons. First, very few
opponents indicated that their primary reason for opposition was the transit
component. Second, the transit component was not particularly costly in this
situation since much of the system had already been constructed.
4
Many people are not in either of these two groups. More specifically, many are pro-highway
and neutral on transit, pro-transit and neutral on highways, or both pro-transit and prohighway. The first point discussed under the survey indicated that a much larger number of
people are pro-highway than pro-transit.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
81
Evaluation of marketing and communications campaign used by
proponents
It is easy to critique a failed marketing and communications campaign after the
fact. Proponents interviewed did so in this situation and some indicated they
did so during the campaign. Certainly, as discussed previously, some things
could have been done more effectively, such as responding more aggressively
in the newspaper once opposition surfaced. However, given the underlying
problem with the paired measures approach discussed previously, combined
with the lack of data from the Calthorpe study supporting an argument that
congestion on 101 would be significantly reduced in the long term with the
transportation package, some fine-tuning changes in the marketing approach
probably would not have changed the final outcome although they could have
closed the margin of defeat. If proponents truly believe that a different
marketing approach will change the outcome, they would have hired a new
consultant and would be shortly presenting the same package to the voters.
Proponents interviewed indicated that this will not be done.
Summary of lessons learned
• An effective coalition of business people and environmentalists can
be developed through a transportation package that has both a
significant highway and a transit component when representatives
from both groups are involved in developing the package.
• It is difficult to generate support for a package with a significant
increase in a sales tax with a long sunset date. The 20-year tax
period in Sonoma County was extremely long.
• It is difficult to overcome voter distrust without a specific plan,
funds specifically allocated for the plan, or a prior track record of
accomplishing plans within budget.
• Even when effective efforts are established to minimize opposition,
such opposition is likely to develop, sometimes from surprising
sources. Contingency plans are needed for such opposition.
• Providing small benefits to some geographic segments seems to
have little impact on voter tendencies.
CASE STUDY III
Seattle: Funding for regional public transit (1996)
In November of 1996, voters in the Puget Sound region of Washington state
passed a $3.9 billion tax increase for the funding of a comprehensive, multimodal, regional, public transit system. The initiative included funding for
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
82
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
transit improvements, including light rail and commuter rail, as well as
expanded express bus service, and high occupancy vehicle (HOV) lanes for
existing freeways. Funding for these improvements was culled from increases
in local sales and motor vehicle excise tax rates. Despite the fact that a similar
initiative had failed the previous year, the 1996 plan passed by a substantial
majority (56.6% in favor regionwide), including majorities in each of the three
counties included in the Puget Sound Region. The 1996 ballot measure
represented the largest transportation tax and public works project ever
approved by popular vote in the state’s history.
This case represents an example of how public officials can learn from past
mistakes and refashion an initiative to maximize public support. It also
provides evidence that strategic use of marketing techniques can help buttress
an effective campaign and confirms the strategic importance of building
effective political coalitions for public transportation.
Among the individuals interviewed for this case were the staff from the local
transportation agency, several consultants associated with the campaign,
Chamber of Commerce staff, and individuals associated with an earlier,
unsuccessful transportation tax measure. Key documents consulted included a
report issued by the local transportation agency (Central Puget Sound Regional
Transit Authority 1997), one issued by the campaign’s consultants (Gogerty
and Stark 1997), and coverage from the Seattle Times.
Background: transportation and traffic
Regional transit has been a concern for the Seattle area since at least the late
1960s. Citizen frustration with growing congestion, fueled by urban sprawl,
has long been present in the area, although comprehensive attempts have
typically failed. Survey data concerning citizen attitudes toward transportation
issues are not readily available but the sources consulted for this review
unanimously support the conclusion that most, if not nearly all, area residents
were very concerned about traffic and congestion and their impact on the
livability of the Seattle area. Empirical data suggests that this concern was
legitimate.
While Seattle consistently tops listings of the nation's most-livable cities, it has
also regularly placed high in surveys of cities with the worst congestion
problems. In 1996, Seattle had the third-highest average round-trip commuting
time: 60.3 minutes (Seattle Times November 1, 1996). The Texas
Transportation Institute ranked Seattle as the sixth most congested city in a
recent ranking of 50 urban areas nationwide.
The population has increased from 1.8 million in 1970 to 2.8 million in 1999 in
King, Pierce, and Snohomish counties, which make up the Central Puget
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
83
Sound Regional Transit Authority (RTA) service area. Residents of the region
are living farther and farther away from where they work. In 1980, 11.6
percent of employees in the region worked in one county while living in
another; in 1990, that number had increased to 15 percent. Many households
have two people working outside the home, meaning more trips per household.
From 1971 to 1987, the average number of daily trips per household increased
28 percent.
A great deal of the congestion is associated with commuting to the workplace
and more and more commuters are driving to work alone. According to the
1990 Census, the proportion of workers in the region who drove alone to work
increased from 64 percent in 1980 to 73 percent in 1990. While total daily
transit trips rose between 1961 and 1990, the proportion of transit trips
compared with all trips dropped from 5.2 percent to 3.3 percent. This figure,
however, obscures the fact that transit is a key component of regional
transportation.
The state Department of Transportation found that vehicle miles traveled
increased by more than 80 percent between 1981 and 1991, from 30 million to
55.2 million. During the same period, the region’s population grew by just over
20 percent, indicating that traffic grew at nearly four times the rate of growth in
the population. The traffic growth resulted more from the increasing number of
miles traveled by each person than from the increasing number of people or
jobs, the study concluded.
Available data suggest that the traffic and congestion woes of the area will
worsen in the future. Since 1990 alone, the average number of automobile trips
has doubled (Newsweek July 19, 1999, p.25). The population in King, Pierce,
Snohomish, and Kitsap counties is expected to increase from about 2.7 million
in 1990 to 4.1 million in 2020. The number of jobs in those counties is
projected to grow during the same period from 1.4 million to 2.2 million.
Prior to the passage of the 1996 transportation initiative, transportation in the
Puget Sound region was essentially limited to buses operated by various
agencies throughout the area and, of course, individual automobiles. At least
three ambitious mass transit plans failed to receive adequate support from
voters over the thirty-year period preceding the passage of the 1996 ballot
measure.
Background: political history
The primary entity with decision-making authority for regional transportation
in the Puget Sound area is the Central Puget Sound Regional Transit Authority
(RTA). Creation of regional transit authorities was enabled in 1992 by the state
legislature, partially in response to the inability of its predecessors to conduct
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
84
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
effective regional transportation planning. In 1993, county councils in King,
Pierce, and Snohomish authorized creation of the RTA. RTA board members
and staff are the primary agents responsible for the transportation plans
recounted in this chapter.
These three counties represent a large, disparate portion of northwestern
Washington, including the cities of Seattle, Redmond, and Bellevue (King
County), Tacoma (Pierce County), and Everett (Snohomish County). Many
smaller suburban and exurban communities fall within the three counties that
comprise the district.
The RTA board is comprised of 18 members, including the state secretary of
transportation (ex officio) and 17 locally elected officials nominated by each of
the three county’s executive official and confirmed by the respective county
councils. Each board position represents 145,000 voters, and one-half of the
board are also members of local public transit agency boards. Thus, although
the Board is tied to the political process, it is not directly elected and is closely
associated with public transit agencies.
The enabling legislation that created the RTA charged it with the duty of
creating and presenting to voters a regional plan within two years. This placed
enormous pressure on the RTA to devise a plan in a limited time period. The
resulting plan was an extremely ambitious one: in October of 1994, the RTA
Board submitted a $6.7 billion, 16-year, multi-modal transit proposal to be
voted on in March of the following year. Only two of the 18 RTA Board
members opposed the plan, including the mayor of the district’s most northern
community, Everett.
According to interviews conducted for this report, this plan was necessarily
assembled in haste in order to meet the legislature’s mandate. The eventual
failure of the plan to obtain the necessary majority of votes may be attributed
to a variety of factors. This report focuses on a subsequent proposal but the fate
of the 1995 proposal is instructive in a number of ways. A widely held view
among those close to the passage of the subsequent measure is that transit
proponents learned from the mistakes associated with the earlier project.
Among the circumstances associated with the failure of the 1995 proposal are
the following:
1) A lack of lead time to conduct campaign. The proposal was required by law
to be approved by each county. Approval was not achieved until approximately
December of 1994, leaving proponents only three months before the March,
1995 election to advance their cause. Opposition to the plan was well
organized, and preceded formal completion of the ballot measure
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
85
(2) The perceived “Seattle-centric” nature of proposal. Many of the proposed
improvements involved projects in and around the downtown core of Seattle,
including an expensive rail tunnel. The light rail line proposal stopped two
miles south of Everett.
(3) A special election ballot. In order to meet the legislative mandate for a vote
on a regional transit plan within two years, RTA was forced to submit the
proposal for a special, off-year election. Such elections are thought to attract
lower turnout and higher percentages of older or tax-resistant voters.
(4) An under-funded campaign in favor of passage. Proponents budgeted a
$1.2 million campaign, but were only able to raise a little more than half of that
amount.
(5) Lingering effects of economic recession. The economy of the region was
threatened by a coincidental announcement from the largest employer, the
Boeing Company, that it would lay off 6,700 workers. Much of the rest of the
local economy was also down.
One other aspect of the campaign warrants special emphasis. The election for
the 1995 proposal was a special one, with no other elections measures
appearing on the ballot. Some scholars have argued that the electorate for
special elections, nearly always smaller, is more fiscally conservative than that
associated with general elections, particularly presidential ones. For that
reason, some observers believe that the 1995 ballot measure would have
passed had it been offered during a general election.
During the brief campaign, opponents of the measure attempted to depict the
plan as elitist, a creation of “top-down” bureaucrats, out of touch with the local
citizenry. Some observers feel that the opposition successfully defined the
issues and controlled the campaign, focusing attention to the details of the plan
rather than the urgency of the need for a regional approach to transportation
issues. In any case, the negative synergy created by these factors was
apparently sufficient to doom the proposal. On March 15, 1995, voters in the
RTA district rejected it by a vote of 53 percent to 47 percent. Voters from
outside of Seattle were much more likely to oppose the plan. In northern
Snohomish county, opponents outvoted supporters by 65 percent to 35 percent.
Developing a transportation coalition
The failure of the 1995 ballot initiative notwithstanding, support among key
business and political institutions was, by all counts, fairly strong if not
manifest in financial support for campaigning. Organized opposition to the
measure was led, at least nominally, by narrowly focused interest groups in
communities furthest from the center of the region, downtown Seattle.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
86
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
However, although it did little or nothing to oppose the plan, neither was the
greater business community an enthusiastic supporter of the plan. Some
funding to oppose the 1995 plan was contributed by pro-highway and
environmental groups. As proponents of a second initiative planned their next
move, efforts were made to court leaders of these groups.
Perhaps of greater significance was the marked increase of support by the
Boeing Company, the largest employer of the region. This development was
probably not solely due to efforts by supporters of a transportation plan;
instead, it was apparently the fortuitous result of the arrival of a new CEO at
Boeing (Phillip Condit) who took a more active interest in public affairs than
his predecessor. Boeing’s importance to the regional political landscape is
even greater than that suggested by its corporate size. Traditionally, the rest of
the business community looks to Boeing for leadership, and tends to scale
donations to political campaigns according to that offered by Boeing.
Creating a transportation initiative
Transit advocates used a multi-faceted strategy to create a new transportation
initiative that could attract majority support from the region’s voters, including
the following components.
• Dividing the region into five areas, in order to create a plan that
would have more widespread support
• Extensive public outreach
• Initial release of “guiding principals,” rather than specific plan
Disaggregation of service area. Faced with the perception, perhaps valid, that
the original (1995) regional transportation was too “Seattle-centric,” RTC
decided to divide the region into five sub-areas: Snohomish County, Seattle,
Pierce County, East King County, and South King County. Each sub-area
would, in effect, help fashion and submit its own plan and the consequent set
of plans would be reformulated into a regionwide plan. This approach,
although it risked creation of an incoherent regional transportation strategy,
helped to ensure that the emergent regional plan would reflect the needs and
interests of the entire constituency. Dividing the region in this way implicitly
acknowledged that passing a transportation initiative “boils down to local
politics.” As the plan for the 1996 ballot initiative took shape, proponents were
able to demonstrate to each part of the region how many resources would be
brought to bear on improving transportation in their area.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
87
Public Outreach
Concomitantly, the RTA created the Regional Outreach Committee to help
ensure that a new ballot measure would appeal to the entire RTA service area.
Members of this 15-member group represented each of the five subareas and
were tapped from existing transportation boards and commissions. The group
met regularly, with meetings open to the public and news media. Informal
meetings with constituent groups, city councils, and other community leaders
were also conducted throughout the sub-areas identified by RTA. Opponents to
the 1995 plan were also invited to participate in various outreach events, some
organized by independent groups. Over the course of the year, thousands of
public outreach meetings were conducted.
Release of “guiding principles.”
Rather than merely scale down the spending called for by the rather massive
1995 plan, the RTA Board endeavored to commit itself to a fresh approach.
Before it set to work on the specifics of the new plan, the board released a set
of “guiding principles” that would oblige it to create a significantly different
plan. Of paramount significance, the board acknowledged that enhanced bus
service would have to be part of the new plan, since rail transit could not
practically serve all suburban areas of the service district. Among the
principles the RTA Board committed to were:
1) The new proposal would be regional, yet would address local needs in each
of the five sub-areas.
2) Spending in the sub-areas would correspond to the taxes raised in each.
3) The proposal would create high capacity transit with a multi-modal
approach, including rail, bus, access ramps for high occupancy vehicle lanes,
and general traffic improvements.
In sum, the board’s approach to transportation planning was significantly
different than that associated with the 1995 plan. It reflected the political and
fiscal realities of the time. By publicly adhering to these principles, the board
was much more likely to create a plan that would attract majority support
regionwide. Although some public opinion polling was used to determine the
overall level of support for a new initiative, yielding encouraging figures, the
details of the plan itself were the result of public outreach in light of the
“guiding principles” listed above. More intensive use of research techniques
did not begin until the new proposal had solidified.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
88
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
Overview of transportation initiative
After consideration of the various forms of public outreach and the “guiding
principals” the RTA Board had established for itself, a draft plan was released
in March of 1996. Release of the proposal inevitably fueled more intense
debate about the details of the final plan among the various groups and
interests whose support was important to the success of the final campaign.
With some minor changes the final plan adopted by the RTA board two
months later (May 1996). Of the 18 board members, only two King County
Council members (from the suburban areas east of Seattle) voted against the
plan. The board voted to submit the measure for the November 1996 ballot,
which allowed five months for the campaign to pass it.
The proposal included the following major components.
• A ten-year, $3.9 billion regional transportation plan funded with
sales and motor vehicle excise tax increases
• Light rail in the region’s densest areas and commuter rail using
existing railroad tracks in other areas
•
•
•
•
Enhanced access to HOV lanes
Regional express bus service
“Seamless” fare integration between regional transit components nd
An independent citizen review panel to oversee RTA and
construction of the system
Compared to the 1995 plan, this proposal was much more modest in scale.
Major differences between the two plans are summarized below. The
difference in cost between the two proposals is primarily the result of a less
elaborate commuter and light rail system in the 1996 proposal. By substituting
less expensive alternatives (bus and HOV access ramp construction), the less
costly 1996 plan appears to directly serve more of the RTA region than did the
1995 proposal.
Clearly, the 1996 package was intended to address the perceived political
weaknesses of the 1995 measure. For example, it eliminated rail transit in East
King county, where opponents had criticized the 1995 plan. The addition of
HOV lanes would help make the plan more attractive to those who believed
that the 1995 proposal would do little to alleviate traffic congestion. Light rail
service from between Tacoma and Seattle was eliminated in favor of a short
link between central Tacoma and the Tacoma Dome commuter rail station.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
89
Table 3-1. RTA Proposal
1995 Proposal
1996 Proposal
Cost
$6.7 billion
$3.9 billion
Life span
16 years
10 years
Commuter Rail
Yes
Yes
Light Rail
Yes
Yes
Express Bus
No
Yes
HOV Access Ramps
No
Yes
Devising and implementing a marketing and campaign strategy
In May of 1996, the campaign committee hired the polling firm of Evans and
McDonough to conduct a public opinion survey about issues associated with
the new ballot measure. Results from the poll indicated that a majority of
respondents were highly concerned with the issue of traffic and congestion and
were supportive of a publicly funded solution. More specifically, the research
showed that about three out of four prospective voters thought “traffic/
transportation was the most important problem facing the region and that a
public solution should be a high priority.” Moreover, 58% of those polled
would vote in favor of “a public solution.” However, results from the poll also
suggested that support for public funding decreased when focus was applied to
the details of the plan.
With these results in mind, the campaign committee hired the consulting firm
of Gogerty and Stark to conduct the campaign, including supporting research
activities. Although much of their more recent work has been for private sector
concerns, Gogerty and Stark had been conducting campaigns since 1968. They
were assisted by a media consultant who created the specific advertisements
used by the campaign.
The general research strategy used by Gogerty and Stark was to use focus
groups as a cost-effective means of understanding voter response to the
campaign, and a series of surveys to monitor voter support. One major
consideration in formulating the campaign was to avoid the perceived mistakes
of the failed 1995 campaign. In the view of the consultants, one of the reasons
that campaign had failed was that opponents were successfully able to make
the details of the plan the campaign’s main issue. However, principals from the
firm also believed that, generally speaking, such initiatives do better after
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
90
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
initial failures. The public seems to be wary of initial proposals, and
collectively “waits for a better deal” that comes with subsequent proposals.
Focus groups were the linchpin research activity for the campaign. Through a
series of focus groups, conducted with prospective voters from various parts of
the region, Gogerty and Stark were able to fashion what turned out to be a
successful campaign strategy. The insights from focus groups were gathered as
follows:
1) Participants were asked about the most important problem facing the Seattle
area. In each group, an average of 8 out of 10 respondents replied
“transportation.” In every group except one, a majority “passed” the actual
ballot language.
2) Respondents were then shown the details of the 1996 regional transportation
proposal in the form of informational brochures and diagrams and asked to
discuss it. As the details of the plan became known, the number of questions
and concerns increased. In most of the groups, within half an hour, most
participants moved from support for the plan to opposition because they
learned it did not benefit them individually.
3) Additional arguments in favor of the measure (such as, “It’s time to do
something!”) left a positive impression with participants, but did not affect
their support for their measure. On the other hand, exposure to opposition’s
arguments against the plan (e.g., “The plan won’t reduce congestion”) seemed
to cause respondents to “step back and look at the whole picture” and reignited
support for it.
From this pattern of responses, Gogerty and Stark formulated a “counterintuitive” communications campaign for the measure. Rather than emphasizing
and extolling the virtues of the details of the plan, the strategy was based on
drawing attention to the opposition and its apparently overplayed message.
The campaign would actually use the opposition’s arguments against
themselves. This approach had the secondary benefit of not allowing the
opposition to focus debate back on the details of the plan, which had been part
of the downfall of the 1995 plan.
The “strategic communications plan” devised by Gogerty and Stark reflected
the results of the focus group research along with a media strategy that would
maximize voter contact. According to Gogerty and Stark, the objectives of the
plan were:
1) Inoculate against the opposition’s anti-arguments by presenting them to
voters first and countering with a positive theme.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
91
2) Explain the general merits of the proposal without getting buried in the
details.
3) Inform voters that the opposition had no plan for an alternative solution.
4) Compare the RTA plan to the “do-nothing” alternative.
5) Ask the voters the question that the proposal answered: “Do you want to get
started on solving our region’s transportation problem?”
6) Establish the message that, “It’s time to do something,” and that, “The RTA
plan is the right first step in solving the regions transportation problem.”
Proponents for the measure had raised $850,000, of which 80 percent was
allocated to citizen contact, as indicated in table 3-2:
Table 3-2. PR Funding
Expenditure
Category
Television ads
Direct Mail
Leaflets
Amount
% of total
Units purchased
$549,000
81.5%
2,000 gross
rating points
$91,000
13.5%
169,000 pieces
$4,000
.5%
15,000 pieces
Radio
Advertising
$16,000
2.3%
4 stations; 6-8
times per day; 5
days
Yard Signs
$13,000
1.9%
5,000 signs
The conscious strategy was to focus on mass media advertising, particularly
television, to create an “impact campaign.” This decision was apparently based
on the corporate philosophy at Gogerty and Stark, which believes that it is
nearly impossible to win this type of campaign without mass media. The funds
budgeted to television ads were intended to ensure that each member of the
target audience of registered voters saw an ad 21 times.
None of the ads or other media used in the campaign were targeted to specific
demographic groups or regions, although survey research had suggested that
women were more likely to support the measure and that opposition was
stronger in the further reaches of the region. Instead, the campaign sought to
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
92
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
saturate the public with the pre-emptive attack on the opposition’s lack of a
viable alternative and its own positive message, “It’s about time.”
Opposition efforts
Opponents of the measure again raised campaign funds from road construction
and suburban real estate interests, but were much less successful in doing so
than for the previous (1995) campaign. They raised several hundred thousand
dollars less than the RTA campaign. Using the acronym “COST” (Citizens
Opposed to Sitting in Traffic), the opposition apparently played into the hands
of the Gogerty and Stark strategy by raising the same issues, and the same
negativism, that had been so successful for them in the earlier campaign.
Unlike the RTA campaign, the opposition relied more heavily on direct mail
and radio spots, presumably because that they lacked the funds to match the
RTA’s television ad blitz. As in 1995, the local news media gave significant
publicity to the opposition, but without the same effect. However, two local
newspapers endorsed the measure.
Outcome evaluation
The election occurred on November 5, 1996, and the measure was adopted by
a sizeable 56.5 percent majority. Perhaps more impressively, the measure won
a majority of votes from all three counties in the RTA service area, whereas its
predecessor had failed in the two suburban counties. How can the success of
the measure best be understood?
First, it is possible that the success of the 1996 initiative can be nearly entirely
explained by the fact it was contested during a high turnout Presidential
election, whereas its predecessor lost during a special election. Given the
insights gained by research during the 1996 campaign, which revealed that
focus on proposal details was generally detrimental to its support, this
interpretation seems even more plausible. During the special 1995 election, the
local news media and proposal opponents drew attention to debate of proposal
specifics in the absence of other elections and issues. The successful proposal
received less attention of this sort.
Second, while the first proposal was in some ways the victim of unfortunate
circumstances, the successful measure benefited from the opposite. The arrival
of a new, more public affairs-oriented CEO at the Boeing Corporation, as well
as the general upswing of the local economy, created a more favorable
environment for a transportation plan of this scale.
That being said, the value of “lessons learned” from the unsuccessful 1995
proposal cannot be overemphasized, and is perhaps the single most important
factor in accounting for the success of the 1996 ballot measure. No single,
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
93
critical “lesson” can be identified, but proponents clearly gained from the
hindsight of the 1995 campaign. Among these lessons were the following:
• The need for a longer lead-in time for the campaign, in order to
raise adequate funds and plan.
• The need to make the plan attractive to all parts of the region, and to
lessen the perceived Seattle bias of the 1995 proposal.
• The need to move the focus of the campaign from a debate of the
proposal specifics to the general need for a public response to the
region’s transportation problem.
Of equal importance, at least in the minds of RTA officials, was the rethinking
of the planning process that anchored the creation of the successful ballot
measure. Rather than presenting the public with a plan, the RTA made
significant efforts to conduct public outreach and solicited input from all areas
within the region. The move was apparently successful in reshaping public
perceptions about the nature of project planning at RTA. Concomitantly, the
agency’s decision to recast the region into five sub-areas helped ensure that the
resulting plan would be perceived as fair to each.
Finally, the strategic approach of Gogerty and Stark was apparently effective.
Focusing the campaign on the opponent’s failure to propose a reasonable
alternative to RTA’s plan made it difficult for the opposition to repeat its
attacks on the RTA’s new proposal. Interestingly, the campaign strategy did
not, for the most, rely upon demographic analysis or other forms of market
segmentation, instead a general mass media approach proved successful.
The Seattle experience should provide encouragement to localities that have
suffered setbacks in passing transportation initiatives. Careful consideration of
past defeats can help ensure the success of future efforts.
CASE STUDY IV
Denver Metropolitan Region: Defeat of guide the ride (1997)
Overview
Guide the Ride (GTR) consisted largely of a proposal for light and heavy rail
lines to be constructed along four or five corridors in the Denver Metropolitan
Region, some feeder bus routes, and unspecified transit modifications on two
or three additional corridors. All the transit improvements covered by GTR
were within the Regional Transportation District (RTD). Prior to GTR, the
RTD district received .006 of sales tax revenues within the RTD for the transit
system. The proposition on the ballot in November of 1997 asked voters to
approve a sales tax increase of .004 to fund GTR. No specific sunset date was
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
94
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
established for the tax. The sales tax increase would be reduced from .004 to
.002 once the bonds were retired. Different bond retirement dates were
identified, with the earliest date being 2020 and the latest being 2035 (Denver
Post October 17, 1997).
GTR was rejected by a vote of 222,163 (57.8%) opposed and 162,196 (42.2%)
in favor. About 29 percent of registered voters voted on the GTR proposition
(Ciruli 1998 p.6). Although the people interviewed were not in agreement on
the most important reasons for the measure failing, many indicated that the
transportation package was too large and had components that were not
specific enough. The lead opponent also developed and implemented a very
effective campaign with very limited money. He was able to get a divided
board to take actions that generated a significant amount of negative publicity
in the newspapers about GTR, and generated voter distrust of the agency that
would need to implement GTR.
The information pertaining to GTR was developed largely through personal
and telephone interviews plus some documents, particularly newspaper articles
from the Denver Post. The people interviewed included editorialists from local
newspapers, board members (which included opponents of GTR) and
managers from the RTD, the campaign’s political consultant, and other
knowledgeable observers and participants.
Background: the RTD
The RTD includes parts of six counties and covers a large geographic area.
Boulder, which is approximately 25 miles from Denver, is in the northwest
part of the district. About half of Douglas County, including parts
approximately 30 miles southwest of Denver is included.
The RTD Board consists of 15 elected members. Even though it uses the word
“transportation” rather than “transit,” it is responsible only for transit. Each
member represents a designated district and each district has approximately the
same population. A nonpartisan election process is used. Elections are often
not strongly contested and some representatives have been elected without
opposition. Voters often know little about the candidates and they tend to
appear at the end of the ballot during U.S. presidential elections. These
circumstances can result in some surprising representatives. For example,
Boulder, which has rather liberal voting tendencies, elected Jon Caldara, who
was the leader of the opposition, to the board in 1994. Two people interviewed
indicated that Caldara’s views were not well known to the voting public in
Boulder prior to his election.
Members on the board during the time of the GTR vote were elected in 1994.
Proponents interviewed acknowledged that they did not work aggressively in
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
95
this critical election to recruit and elect candidates likely to be supportive of
GTR. On the other hand, the Independence Institute, a free-market think tank
that opposes light rail, worked hard to recruit and elect like-minded candidates
to the board in 1994. In the absence of an organized effort by transit advocates,
the efforts of the Independence Institute were successful. Even though the
population generally viewed traffic congestion as a bad problem that was
getting worse and had positive attitudes towards light rail, five light rail
opponents were elected to the Board in 1994. After the defeat of GTR, an
organization entitled Metro Transit, affiliated with the Chamber of Commerce,
was formed to recruit and elect pro-transit people in the 1998 election. These
results appear to have been successful as five out of the seven less-thanenthusiastic supporters of GTR were replaced by pro-transit representatives in
1998.
Background: the transportation system and voter attitudes
Prior to GTR, the transit system consisted of buses, including some express
buses using expressways and freeways, and one 5.5 mile light rail system in
Denver that opened shortly before GTR. In addition, a light rail system along
the southwest corridor had been started prior to GTR without the need of
additional taxes. An attempt was made in 1980 to raise sales taxes to fund a
light rail system, but it was defeated (Ciruli 1998, p.6).
Only limited improvements in the highway system were made in the 1980s and
1990s. The only change was the addition of HOV lanes on I-25, which runs
due north from Denver, and on C-470, which runs in an east-west direction
south of Denver. The metropolitan area had a designated non-attainment status
until the mid-1990s, which precluded significant highway improvements.
All people interviewed for this report indicated that the public viewed traffic
congestion as a high priority problem. In 1995 commute drive times along the
approximately 20-mile I-25 corridor, which is the most congested corridor,
were about 45 minutes in each direction without unusual incidents that slowed
down traffic. This time was projected to increase to 66 minutes by 2020 (Carter
and Burgess 1997 p.4-32). In addition, surveys indicated that people generally
supported light rail (Carter and Burgess 1997 p.3-8). A poll taken in May,
1997, found that 59 percent of those surveyed supported GTR even when they
were informed of its approximate construction costs (Ciruli 1998 p.10).
In general, Colorado voters seem to be rather middle of the road. Roy Romer, a
Democrat, was governor from 1986 until 1998 when a conservative, anti-tax,
Republican, was elected by a very small percentage. Statewide tax increase
elections have not fared well in the 1990s. For example, then-Governor Romer
presented a one-cent sales tax initiative to the voters in 1992 for K-12
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
96
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
education which was defeated by 10 percentage points. Local level tax issues,
however, were often passed during the 1990s. During that time there were 10
proposals that involved overrides of the Taxpayer’s Bill of Rights (TABOR).
Although these involved permission to spend tax revenue already collected,
rather than to increase taxes, 9 of the 10 override elections passed (Ciruli 1998
p.2).
The transportation package and the process used to determine it
The package involved a comprehensive hub-and-spoke rail system with all
except one rail line converging in downtown Denver. Since major investment
studies had not been completed on two northern corridors, the improvements in
these corridors were not specified in the package. This lack of specificity was
one of the major criticisms of GTR advanced by opponents. The package also
did not involve any increase in highway capacity, except for the possibility of
HOV lanes along the northern corridors.
Different estimates of the costs existed and this was attacked by the opponents.
The cost estimate advanced by proponents was about $8 billion, about half of
this figure involving construction costs. The remaining half consisted of
financing and operating costs. The lead opponent, Jon Caldara, who also was
on the RTD Board, used a figure of 16 billion dollars. He was successful in
getting the Board to accept, by an 8 to 6 vote, the 16 billion dollar figure as the
official estimate of the Board.
Regardless of which figure was used, the costs were quite high. To put this in
perspective, GTR involved an increase in the sales tax of .004, from .006 to
.01, that would last until the debt was fully repaid, which was estimated to be
in 2020. Once the debt was paid, the tax would be reduced by .002 to .008.
Thus, although no specific sunset date existed, an estimated date of about 20
years existed on half of the .004 sales tax increase. The remaining half of the
tax increase, .002, was a permanent sales tax increase. The combination of this
increase in the sales tax and no specified sunset date made this package
significantly more costly than the other cases examined in this project. The
lack of a specific sunset date may have combined with different cost estimates
to reinforce the perception that voters would be writing a blank check.
Interviews indicated that the decision to focus almost exclusively on transit
was largely the result of jurisdictional responsibilities. The RTD Board was
only responsible for transit while the state had the primary responsibility for
highways and the non-attainment status of the region precluded significant
highway improvements during the time frame.
The decision to have a comprehensive but expensive package was based on a
variety of factors. One critical factor was the perceived need to offer something
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
97
for each geographic region. As one proponent interviewed indicated, the
package ended up being a “Christmas tree.” As one example of this desire to
offer something for each corridor, GTR included unspecified transit
improvements on two northern corridors even though no corridor studies had
been completed for them. The board also realized that a systematic and
comprehensive package would be cheaper in the long-run than a more
incremental approach and also would generate a larger increase in ridership,
since people take trips across corridors. Although it was not mentioned by any
of the people interviewed, a concern might have existed that an incoming
governor was more likely to be less supportive of transit than Romer was. This
scenario did occur with the election of the new governor in 1998.
The transportation package was selected by RTD staff using major investment
studies. A number of public hearings were held as part of the Management
Information Systems (MIS) process and the transportation package was
expanded by approximately $1 billion.
As a result of these public hearings. The board voted 9-6 to approve the
transportation package that was presented to the voters.
Given the lack of a strong consensus on the board regarding the GTR package,
the proponents could have tried to modify the package to gain more Board
support. However, it is unlikely that more than one additional vote of support
could have been generated. As mentioned previously, according to multiple
people interviewed, five board members elected in 1994 were opposed to
increased taxes and also opposed to light rail. On the other hand, seven or eight
board members were strong proponents of light and heavy rail.0 Given this
polarization between these two groups, obtaining consensus through a
compromise was virtually impossible. Certainly a package involving a
combination of transit and increases in highway capacity would not have
created a stronger consensus.
It is interesting that surveys of voter preferences for different transportation
options were not considered in formulating the transportation package.
According to the general manager of the RTD, some general polling was
completed prior to putting the package together and the results indicated a
general approval of light rail. However, another source indicated the RTD had
a survey conducted in 1994 testing tax tolerance levels and these results
indicated that a .008 tax was the highest level the public would support (Ciruli
1998 p.7). The GTR package, consisting of a .01 sales tax (a base of .006 prior
to GTR plus a proposed increase of .004 to fund GTR), was greater than the
.008 level suggested by this RTD survey. The general manager may not have
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
98
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
been aware of this 1994 study since he did not start his work with the RTD
until 1995. At any rate, the results of the study were not heeded.
Proponents acknowledged, after the fact, that chances of voter approval would
have been greater with a different package. The proponents interviewed
indicated that it was a mistake to put forward a package that did not have a
defined solution along two northern corridors. None of these proponents,
however, felt strongly that voter approval would have occurred with a cheaper
package, nor did any indicate that they should have presented a package
involving a mix of highway capacity improvements and light rail.
According to the general manager of the RTD, voters in the district will be
asked to vote on a transportation package in the November 1999 election, and
this package will be significantly different than the package with GTR. The
differences suggest that proponents have used the failure of GTR to
reformulate the transportation package. This $800,000 transit package does not
involve any additional taxes and it involves only one additional light rail line
along the southeast corridor. General agreement exists that this is the most
congested remaining corridor. In addition, the governor has put a statewide
$2.3 million dollar package for highway improvements with no new taxes on
the November 1999 ballot. This package includes lane additions to I-25 along
the same southeast corridor. Although these are two separate proposals, the
governor’s office is working closely with the RTD Board to obtain passage of
both propositions. Thus, for all practical purposes, voters are being asked to
approve a package consisting of both highway and transit improvements along
the SE corridor. A poll taken in June 1999 generated voter approval ratings of
over 70 percent with each of these two packages and these approval ratings are
significantly higher than GTR received at a similar point prior to the election
(Denver Post June 22, 1999 p.A-1).
The campaign strategy used by opponents of GTR
A summary
The campaign to oppose GTR was organized and implemented by little more
than one person with no political consultants, no voter research, and with very
little money. This person was Jon Caldara, who was also on the RTD Board.
The proponents, as well as the three newspaper people interviewed all agree
that Caldara was very effective as an opposition leader. He worked tirelessly to
defeat GTR, and he is a persuasive individual.
Caldara’s message focused on the high costs, the minimal reduction in traffic
congestion, and the vagueness of GTR along a couple of corridors. According
to proponents, he did not present an alternative transportation package to the
public. The basic challenge he faced was how to present this message with
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
99
very little money available for paid advertising. His strategy consisted of
establishing credibility for his arguments by getting the RTD Board to vote for
propositions he presented to the board and obtaining frequent newspaper
coverage prior to the election with an emphasis on characteristics of GTR that
would work in his favor even with balanced coverage. These two major
components were interrelated. More specifically, he was correct in assuming
that the two major newspapers would write articles about the board’s actions,
particularly given that the election took place in an odd-numbered year in
which GTR essentially was the only political issue on the ballot.5 Thus, by
using his position on the board to bring controversial proposals to it for a vote
that could generate voter doubt about GTR, Caldara was able to generate many
articles in the two newspapers about the board’s actions. To increase the
chances that the newspapers would provide extensive coverage of the Board’s
actions, Caldara contacted the transportation news writers of each paper
frequently. In contrast, according to one of the reporters, proponents rarely
initiated communications with the newspaper to obtain coverage. Each of these
two major components of Caldara’s strategy is discussed in some detail below,
followed by a discussion of some additional elements of his strategy.
Influencing the Board to take two actions
Caldara had to persuade two board members to switch sides, since he started
with a coalition of six members and he needed eight votes. Without getting into
a detailed discussion of possible reasons for two members switching sides, it
should be noted that interviewees thought the defections could have been
prevented if proponents on the board had made a more concerted effort to
maintain their majority. More important, as stated previously, Caldara could
have been prevented from getting a majority if proponents had made a more
concerted longer-term effort to prevent the five libertarians from being elected
in the 1994 election.
The first proposal that Caldara brought to the board was entitled “Guide the
Ride is Not for Sale.”
“Let it be resolved, that in the event the “Guide the Ride” sales tax
referendum wins passage, the district will not enter into any
financial or contractual relationship, until the year 2015 with any
individual or entity who donates more than $100.00 after August 1,
1997, in a contribution or a contribution in kind as defined by
5
According to two newspaper people interviewed, one from each of the two major
newspapers, both papers would have had significantly less coverage of GTR if there had been
other political issues on the ballot.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
100
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
Colorado Statute 1-45-103, directly or indirectly, to any campaign
whose purpose was to effect the outcome of the referendum.”
This resolution passed by a vote of 8 to 6 on August 19, 1997. It was declared
unconstitutional prior to the election, appealed, and the appeal upheld the
lower court’s decision. Although this resolution probably hurt the fund raising
efforts of proponents, the main effect may have been the generation of negative
news coverage. A front page news article in the Denver Post focused on the
resolution the day after it passed. The initial article was followed by eight or
ten articles on the same topic in September. These articles were balanced with
quotes from both sides. Caldara, along with the two board members who
shifted, including the board chair, were consistently quoted in support of the
measure and their basic argument was that most of the contributors were
contractors and bond firms which had a vested interest in the measure passing.
They positioned themselves as insuring a fair election. The other side was
basically quoted as saying it was unconstitutional to prevent people with a
vested interest from making contributions. Clearly, the opponents had a
position that the voting public was likely to support. Equally important, voters
may have gotten the impression from this series of articles that groups with a
vested interest in having GTR passed were the only groups that strongly
supported GTR.
In the middle of October, Caldara presented another resolution that required
the RTD to issue a press release announcing the official cost of GTR as
$15.609 billion. In addition, this resolution required the RTD to use a cost of
$199.36 per year for a family of four with the .004 sales tax increase, even
though RTD staff had estimated the cost at $59 per year. This resolution also
passed with eight members supporting it. Again, a series of front page articles
in the Denver Post focused on this cost issue. The articles explained the
different assumptions involved in the conflicting figures. Given the high costs
of GTR with even the lowest cost estimate, this focus on costs also was not
beneficial to proponents.
Newspaper coverage
The impact of these resolutions on voters probably would have been minimal if
the newspapers had not focused extensively on the funding and cost issues. Yet
the newspapers did focus extensively on these issues, others involving GTR. A
database search conducted by one of the newspeople interviewed identified
185 articles that focused either in their entirety or in part on GTR in the Denver
Post, with at least 10 of these were lead front-page articles.
A content analysis of the newspaper articles suggests that the topics were more
favorable to the opponents than to the proponents. Most of the articles dealt
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
101
with the previously mentioned contributions and cost issues. The cost
estimates were presented in virtually every article in the Post. Given that even
the lowest cost estimate was $2.1 billion, combined with an increase over time
in the estimated costs, the focus on costs was beneficial to opponents of GTR.
The articles also focused on the conflict within the RTD Board, particularly
those in the Rocky Mountain News. In contrast, only one article was found that
focused on current levels of congestion with predictions of significantly higher
levels without action being taken. This latter focus certainly would have been
the one most preferred by proponents. The newspaper people who made
decisions on what dimensions of GTR and the board to emphasize were not
attempting to influence voters. Their decisions, however, might have been
different if proponents had made active efforts to obtain more coverage on
aspects of GTR that would have been more favorable for them.
Advertising
According to a newsperson interviewed, Caldara did raise about $45,000,
mostly from anti-tax people and car dealers. He spent money for radio
($17,000) and TV advertising ($14,000) (Denver Post November 5, 1997 p.A1). The slogan of the radio ads was “Guide the Ride sounds good but it’ll put a
big dent in your wallet - not traffic.” (Denver Post October 31, 1997 p.B-4).
One of the proponents remembered the TV advertising of the opponents even
though very little money was spent. He described it as consisting of visuals
comparing congestion on I-25 with and without GTR and the views were
virtually identical. It is interesting that this same person could not remember
the advertising message for the proponents even though they spent a lot more
on advertising.
The description of the opponents’ strategy may have created the impression
that Caldara and other board members were the only public officials who
opposed GTR. This impression would be incorrect. An article in the Post
shortly before the election (November 2, 1997 p.A-1) indicated that 31 public
officials endorsed a no vote on GTR. It is interesting that this same article did
not identify the corresponding number who supported GTR. Prior articles
tended to balance a positive endorsement from an elected official with a
negative endorsement from another elected official. Thus, at least toward the
end of the election, Caldara was far from being the only elected official who
recommended a no vote on GTR. In addition, one of the two major
newspapers, the Rocky Mountain News had a series of editorials
recommending a “no” vote. Their editorial position will be discussed in more
detail under the proponents’ strategy.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
102
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
The campaign strategy used by proponents of GTR
A political consultant was hired around June 1997 to develop a campaign
strategy. Thus, the campaign got off to a late start.
Advertising: Advertising was the major communications component used by
proponents. According to the Denver Post (November 5, 1997 p.A-1),
proponents raised $550,000. The political consultant had developed a budget
of $970,000 so the fund raising campaign did not reach its goal. Caldara’s
resolution preventing organizations that contributed more than $100 from
obtaining any contracts resulting from GTR probably was a critical factor in
the fund raising shortfall. As a result of this shortfall, the amount of money
spent on TV advertising was cut back significantly. The proponents ended up
spending $250,000 on direct mail and $185,000 on radio and television
(Denver Post November 5, 1997 p.A-1).
The major message in the media campaign was that traffic congestion was bad
and getting worse and that it would be lessened for everybody, including
drivers, by taking cars off the roads. For example, a radio ad consisted largely
of testimonials of people currently taking transit. One of the people was quoted
as saying “I find it mind-boggling that it takes an hour to go 15 miles.” (Denver
Post October 16, 1997 p.B-3) Proponents appeared to be reluctant about
presenting numbers to support their message. Although it was never explicitly
stated by any proponents or the political consultant, proponents probably did
not want to encourage media coverage, believing that it would give opponents
a platform, even if the coverage were balanced. This aspect of their strategy
will be discussed in more detail later in this section.
The political consultant thought they could have had a better theme that
grabbed the voters. When asked what such a theme would be, however, he did
not identify one. It is not surprising that he could not develop such a theme
given the nature of GTR. The underlying problem was that MIS studies
suggested very little benefit to continuing drivers with GTR. More specifically,
the key issue here was whether their commute time would be significantly
reduced. According to the MIS studies, commute time would not be
significantly reduced. For example, the MIS study for the southeast corridor,
which is the most congested one, predicted a driving time of 68 minutes along
this corridor with no capacity increase compared to a driving time of 66
minutes with light rail (Carter and Burgess 1997 p.4-28). It was difficult to
make a strong case in advertising when such a case cannot be made
realistically on an issue of critical importance to many voters.
The direct mail was targeted at those who voted in at least two of the three
most recent odd year elections and also voted in the 1996 election. The
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
103
consultant estimated that this campaign reached 68 percent of those who voted
in the 1997 election on GTR. Three mailers were sent to all the targeted voters
and a fourth piece was sent to a subset of the target. Each piece gave a
descriptive overview of GTR, but emphasized the components most directly
relevant to voters in each county. In addition, local testimonials were used that
differed for each county. The first mailer was sent about two weeks prior to the
election although the schedule varied depending on whether voters in a county
could vote by mail.
Publicity
Proponents seemed to hope that news coverage would be minimal. This hope
was reasonable, as such coverage would tend to be balanced and give the
opposition a platform. Unfortunately for the proponents, they could not
realistically prevent news coverage, particularly in a year in which no other
political issues were on the ballot. They neither tried to actively influence the
news coverage nor aggressively respond to it even when it became evident
both that substantial coverage existed and that Caldara was very effective in
this arena. Two proponents were critical of this media approach.
Another relevant component of publicity involved the attempt to obtain
editorial support for GTR. More specifically, the proponents may not have
realized the importance of preventing either of the two major newspapers from
recommending a “no” vote. Given the composition of the editorial board of the
Post, the proponents would have had to be very ineffective to not gain their
editorial support. They were successful in not losing this support. The Rocky
Mountain Times position was more of a challenge. Caldara acknowledged the
importance of obtaining editorial support from the Rocky Mountain News for
his position and he worked hard to obtain their support. They did end up
supporting the opponents, after remaining uncommitted during the first part of
the campaign. The member of the Rocky Mountain News editorial board who
was interviewed indicated that the editorial board was quite undecided about
their position prior to meeting with representatives of both proponents and
opponents. He also indicated that the representatives for the proponents who
appeared before the editorial board did a very poor job of presenting their case.
He thought they basically dealt with platitudes such as it being civically
responsible to support GTR and that they were not prepared to counter charges
of it being too expensive, poorly defined, not having a sunset date, and having
only a limited impact on congestion and pollution. Largely as a result of the
ineffective performance of proponent representatives with the editorial board
of the Rocky Mountain News, the board decided to recommend a “no” vote.
Once they made this decision, they presented a number of editorials with this
recommendation. Their basic reasoning was that GTR was too expensive given
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
104
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
the relatively small impact on congestion and pollution, too vague on cost
estimates, and did not have a reasonably short and specific sunset date. Since
the person interviewed indicated that he is in favor of light rail for the
southeast corridor, the newspaper’s editorial position was not based on
fundamental opposition to transit and light rail.
Results of a post-election survey
Since proponents made a number of mistakes, information from voters is
helpful in identifying the most critical mistakes. Fortunately, in this case, as
well as in the other failure case studied, Sonoma County, opposing voters were
surveyed shortly after the election about the primary reason for their
opposition. The results of a response to an open-ended question are identified
below (Ciruli 1998 p.9).
Plan too vague, don’t know what
it costs/ costs keep changing
33%
Too expensive/too big a
project/taxes too high
31%
Don’t trust board to manage
it/too much board conflict
15%
People won’t use it/light rail
doesn’t work/won’t reduce
congestion/won’t reduce pollution
11%
Our area doesn’t get benefits/
doesn’t help suburbs/our area
doesn’t get light rail
9%
Since all five of the above reasons were identified by opponents and also
identified in a series of editorials in one of the two leading newspapers in the
region, one question is whether voters would have identified these reasons on
their own. Would GTR have been successful if opposition had not developed
or the newspapers had not raised questions about the costs, vagueness, the
board, and the benefits? Although the answer to this hypothetical question
cannot be known with certainty, it appears that the vagueness and the high
costs were readily apparent with even a basic knowledge of the measure’s
characteristics.
As discussed in the prior chapter, previous research has indicated that
successful passage is lessened by both a transportation package consisting only
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
105
of passenger rail and a package that does not provide geographical equity.
Given that these two reasons were fourth and fifth in the percentage of
responses, the relevant question is whether this package would have succeeded
if it had consisted only of light rail lines on one or two corridors. This
incremental approach certainly would have reduced the costs of the package. It
also would have provided a more specific package with more stable cost
estimates. In other words, such an incremental approach would have
eliminated the two top reasons cited by almost two-thirds of the opposing
voters in the survey. If the opposing voters, who selected one of the top two
reasons as their primary reason for opposing GTR, had no additional reasons to
oppose the measure, then such an incremental approach probably would have
succeeded. At least some of the opposing voters, however, who identified one
of the top two reasons as their primary one, also perceived that at least one of
the last three reasons existed. Thus, it is difficult to determine whether an
incremental passenger-rail-only transportation package would have succeeded.
County-by-county voting percentages
It is relevant to compare the county-by-county voting results for several
reasons. First, since congestion was not equal on all corridors in the region, this
comparison can examine the issue of the extent of a correlation between voting
results and the amount of congestion. For example the amount of congestion
was greatest for residents in Douglas and Denver Counties. Second, no
specifically defined transportation projects were identified for either Boulder
or Adams County. Thus, a county-by-county comparison can examine the
amount of correlation between voting results and the vagueness of
transportation projects.
County comparisons in the GTR election in 1997 can be better understood with
a base level for each county. A measure of change in voting results within each
county from an election comparable to GTR provides such an understanding.
Fortunately, such a comparable election exists. A TABOR election override
took place in 1995 that would allow already collected surplus revenues to be
used for transit in the district. The percentage of “yes” votes for each county in
both the 1997 GTR election and the 1995 TABOR override transit election is
identified in Table 3-3, followed by an interpretation (Ciruli 1998 p.6).
The percentage change results indicate that a correlation exists between voting
patterns and a) whether transportation improvements within a county were
clearly specified and b) the extent of congestion in the county. Adams and
Boulder were the two counties in which the transportation improvements were
not clearly specified and the greatest percentage change existed with these two
counties (-23% and –17.3%). The percentage change was the lowest in the two
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
106
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
counties, Denver and Douglas, in which congestion was the worst (-.6% and
–8.4%).
Table 3-3. The percentage of ‘yes’ votes for each county in both the 1997
GTR election and the 1995 TABOR override.
County
95 TABOR
Override
97 GTR
Percentage Change
Adams
41%
23.70%
-17.30%
Arapahoe
55%
44.80%
-10.20%
Boulder
63%
40.00%
-23.00%
Denver
51%
50.40%
-0.60%
Douglas
55%
46.60%
-8.40%
Jefferson
52%
43.30%
-8.70%
Implications
Since GTR lost by a large margin, the lessons learned involve understanding
things that proponents did not do that should have been done, along with things
they did that should not have been done. In general, proponents could have
done a better job of involving both the public and stakeholder groups in the
process of developing a transportation package, developing a specific
transportation package with a reasonably short sunset date with documented
benefits, and planning and conducting an effective marketing and
communications campaign when effective opposition existed.
• When direct elections are used to determine the composition of the
governmental agency responsible for developing the transportation
package, proponents should organize to nominate and elect
representatives likely to be supportive of their agenda. In this case,
five libertarians were elected to the RTD Board in 1994, some with
very little opposition from transit advocates.
• An underlying institutional structure can make it difficult to
develop a transportation package likely to be approved by a
majority of voters. In this case, the RTD Board could not develop a
transportation package with both a highway and a transit
component.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
107
• Funding initiatives with a long or indefinite sunset date are
problematic. This problem is magnified when published cost
estimates keep increasing by a significant amount, the agency
responsible for spending the money is divided and has no prior
experience demonstrating their capability to implement a complex
transportation package within budget, and some components of the
transportation package are not specifically identified. A significant
amount of voter mistrust will tend to exist with this combination of
factors.
• The attempt to provide geographical equity across a large and
heterogeneous region easily can result in a transportation package
that is prohibitively costly, unless energetic efforts are taken to
control costs. Energetic efforts were not taken in this case. As
indicated in the body of the chapter, the RTD Board did not use the
results of a survey, conducted a few years prior to GTR, which
identified a sales tax threshold of .008. The lack of energetic cost
control efforts is also demonstrated by the high cost of this plan. It
was at least $4 billion more than the cost in the Puget Sound region,
even though both regions are comparable in size.
• Relying almost exclusively on corridor studies to develop a
transportation package will tend to result in packages that are less
than optimal from a voter perspective. The RTD Board did not use
any voter research techniques to help develop their transportation
package. Although they had conducted survey research indicating
that voters had positive attitudes towards passenger rail transit and
also were dissatisfied with highway congestion, they did not seem
to use this research to set priorities needed to control costs. If they
had done such research, they probably would have discovered, as
surveys in Sonoma and Santa Clara County did, that voters would
be hesitant to support a very expensive package that consisted
almost exclusively of passenger rail transit.
• Do not develop an expensive transportation package unless corridor
studies indicate that the package will provide significant highway
congestion relief. In this case, opponents used findings from the
corridor studies to support their contention that GTR would have
very little impact on highway driving times during the commute
periods. Proponents could only respond to this argument with
generalizations or statistics not directly relevant, such as the
number of cars taken off the roads. A discerning press can and
usually will identify the flaws with this response.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
108
Case Studies: Campaigns for Transportation Tax Measures in Four Communities
• Although it may be fortuitous for proponents when extensive media
coverage does not exist because opponents have very little money
to spend and when early tracking surveys predict voter approval,
contingency plans are needed when extensive coverage does occur.
One component of this plan involves identifying the specific issues
that are more favorable to proponents than opponents and
proactively encouraging the media to cover these issues.
• Coalitions should be developed while the package is being
formulated by involving relevant groups in the decision-making
process. If more representatives from various geographic segments
had been involved in this process, the problems experienced in
having a package with undefined components on some corridors
would not have existed. In addition, if elements from the business
community had been more enthusiastic about GTR, the proponents
would not have had to rely for funding so extensively on
contractors and lending companies, who were perceived as having a
vested interest in the outcome.
In the next chapter the findings from the preceding case studies are analyzed
and synthesized.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
109
CONCLUDING OBSERVATIONS AND
RECOMMENDATIONS
INTRODUCTION
What lessons may be culled from the case studies presented in the previous
chapter? How can advocates of transportation measures apply these lessons to
their own jurisdictions?
This chapter presents a comparison of the four cases on a number of
characteristics: 1) the nature of coalitions formed, along with the process used
to form these coalitions, 2) the process used to formulate a transportation and
funding package, along with the nature of the package, and 3) the marketing
and communications strategy and tactics used by both proponents and
opponents to try to influence voter approval. The collected characteristics of
each case study are presented in Table 4.1.
A discussion of critical success and failure factors follows Table 4.1. Many,
but not all, of these factors differentiate at least one of the two successful cases,
(Santa Clara County and the Puget Sound region in Washington State) from at
least one of the two unsuccessful ones (Sonoma County and the greater Denver
metropolitan area).
TABLE 4-1. Comparison of Cases on Relevant Characteristics Pertaining
to Coalition Building, a Transportation and Funding Package, and
Marketing and Communications Programs Targeted at Voters
Santa
Clara
Outcome
Won 52%
Seattle
Won 56%
Sonoma
Denver
Lost 48%
Lost 42%
Transportation Proposal Characteristics
Type and amount
of tax increase
General
sales tax
(1/2 cent)
Auto excise
tax (.3%)
General Sales
tax (.4%)
General
sales tax
(1/2 cent)
General
sales tax
(.4%)
Duration
9 years
9 years
20 years
No sunset
Gross revenues
(estimated)
$1.1
billion
$3.6
billion
$1
billion
$8–16
billion
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
110
Concluding Observations and Recommendations
Santa
Clara
Seattle
Sonoma
Denver
Type of election
General
General
General
General
Projects spread
across district
Yes
Yes
Yes
Yes
Multipurpose
projects
Yes
Yes
Yes
No
Campaign Characteristics
Campaign
consultant hired
Yes
Yes
Yes
Yes
Use of surveys (to
design proposal)
Yes
Yes
Yes
No
Use of surveys (for
campaign)
Yes
No
No
Yes
Use of focus
groups (to design
proposal)
Yes
Yes
No
No
Use of focus
groups (for
campaigns)
No
Yes
No
Yes
Received
newspaper
endorsement
Yes
Yes
Yes
Yes and
No
Media coverage
(volume)
Light
Light
Heavy
Heavy
Outreach effort
Yes
Yes
Yes
No
Use of mass media
(by proponents / by
opponents)
Yes
Yes
Yes
Yes
Use of direct mail
(by proponents / by
opponents)
No
No
Yes
Yes
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
Santa
Clara
Seattle
Sonoma
111
Denver
Coalition Characteristics
Unanimous support
from executive
board
Yes
Yes
(virtually)
Yes
No
Organized
opposition
No
Mild
Yes
Yes
Prior defeat
Yes
(1994)
Yes
(1995)
Yes
(1990)
No
Key industry
support
Yes
Yes
Yes
Yes
Environmental
group support
Some
Some
No
Yes
CRITICAL SUCCESS AND FAILURE FACTORS:
TRANSPORTATION AND FUNDING PACKAGE
The analysis begins by examining characteristics related to both the nature of
the transportation projects and the process used to select the projects. Coalition
building implications related to the transportation package also will be
discussed in this section.
Sunset date for the tax
Both Middleton (1998) and Beale et al. (1996) recommend that a reasonably
short sunset date is needed to obtain voter approval of transportation funding
initiatives. The four cases described in this study provide support for this
recommendation.
Each of the four cases relied either primarily (Puget Sound) or entirely on the
sales tax as a funding mechanism (Santa Clara County, Denver, Sonoma
County), with sales tax rates of .004 or .005. However, the two successful
cases both had sunset dates of approximately 10 years, while both unsuccessful
cases had a sunset date of at least 20 years. These results provide strong
support for the hypothesis that sunset dates of ten years or less are needed to
obtain voter support when sales tax rates are approximately .005.
The case analyses also demonstrated the difficulty of designing a
transportation package that has a sufficiently short sunset date. This difficulty
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
112
Concluding Observations and Recommendations
is demonstrated in both of the unsuccessful cases wherein the cost of the final
transportation package presented to voters was significantly higher than the
cost of the initial packages considered. The size and corresponding cost of the
transportation package increased over time because the easiest way of
satisfying different constituencies is to provide additional benefits. For
example, in the Denver case, unspecified projects were added for the two
corridors north of Denver to try to get approval of voters living in counties
along these corridors. These additions increased the cost of the transportation
package while having no apparent positive impact on voting results in these
two counties.
Establishing a budget constraint
One of the key characteristics of a disciplined process to minimize the costs of
a transportation package involves starting with some sort of budget constraint.
Essentially, this was done in both of the successful cases but was not done in
either of the unsuccessful ones. In the Santa Clara County case, the budget
constraint was established through a survey of voter attitudes towards various
combinations of sales tax rates and sunset dates. This research identified a
sunset date threshold at 10 years and a sales tax rate threshold of a half cent. A
budget cap was determined using these results by simply forecasting the sales
tax revenue likely to be raised with a half cent sales tax in nine years.
The RTA in the Puget Sound case did not start with a firm budget cap.
However, they did start with a goal of significantly reducing the cost of the
package from the prior package, which was defeated. In addition, they
appeared to have a process that constrained the package to meet a sunset date
of approximately 10 years.
Identifying priorities
In successful cases, the use of a budget constraint was accompanied by a
systematic process of prioritizing components of a transportation package.
Starting with an appropriate framework may prove useful in helping establish
priorities, as this was done in both successful cases and in the Sonoma County
case.6
The most difficult step in establishing priorities involves prioritizing specific
components of a transportation package. This step should involve using 1)
cost/benefit information based upon corridor studies, 2) voter preferences,
attitudes, and priorities based upon surveys, focus groups, and outreach
6
Although an appropriate underlying framework was established in the Sonoma County case,
this framework was not used to prioritize transportation projects within a budget constraint in
this case.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
113
programs, and 3) input from key interest groups through a citizen advisory
committee.
It is possible that all three of the above sources are part of a successful
campaign, as they are potentially complementary. Corridor studies may prove
insufficient without using at least one of the other two sources. Although these
studies claim to examine cost/benefit ratios, it is very difficult to determine the
added value to the public of a more expensive option that seems to delivers
more benefits compared to a less expensive option with fewer apparent
benefits. The mistake of relying almost exclusively on corridor studies was
demonstrated in the Denver case. Although that campaign did include
extensive public hearings, no systematic voter research was conducted to
identify priorities and comparative perceptions and apparently nothing was
done to integrate interest groups into the planning process through a citizen
task force.
On the other hand, all three sources were used in the successful Santa Clara
County case. Establishing priorities was a significant challenge in this case
because not all viable options could be funded within the budget constraint. In
other words, the final transportation package presented to voters was smaller
than the starting package. One of the specific sources of information about
voter preferences involved a survey that presented respondents with a number
of specific transportation projects and asked for prioritized responses. This
seems to be a very useful tool to use in the prioritization process.
Different approaches were used to reduce the costs of the transportation
package in the two successful cases. The 1996 Puget Sound package
substituted the combination of HOV lanes and express buses for some more
costly passenger rail lines that existed in the 1995 package. An incremental
strategy was used to cut costs in Santa Clara County. Parts of some light rail
lines were eliminated with the understanding that voters would be asked to
approve extensions of these lines in the future. Although this incremental
approach had been used previously in the county with success, it does involve
a risk: it could be positioned by opponents as having a longer sunset date than
it appears since it may be difficult to pull the plug on a partly completed
project. Since opposition was poorly organized in Santa Clara County,
additional research is needed to help determine the significance and magnitude
of this potential risk. On the other hand, voters might prefer the incremental
approach because it allows them to observe whether a first phase of a project is
successful prior to committing to additional phases. Judging the quality of
plans created through such a process is beyond the scope of this research,
however.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
114
Concluding Observations and Recommendations
Geographic equity
Middleton (1998) and Beale et al. (1996) have identified geographical equity
as a critical success factor in obtaining voter approval of transportation funding
initiatives. The case analysis lends support for this conclusion but also
identifies an important risk and challenge in trying to provide meaningful
equity. In each of the two successful cases, concerted and successful efforts
were made to provide transportation projects to each geographic segment
roughly proportional to the sales tax revenues likely to be generated in each
segment. This was a high priority goal in the Puget Sound case since one of the
primary reasons for failure in 1995 was a lack of geographic equity. A
decentralized planning approach was used to accomplish this goal.
On the other hand, geographical equity did not exist with the two failed cases
as well as with the failed 1995 attempt in the Puget Sound case. In each failed
case, some effort was made to achieve such equity but these attempts probably
did more harm than good. In the Denver case, the attempt to include all
relevant corridors resulted in vague plans with some of the corridors where
corridor studies had not been completed. In addition to not gaining support in
these corridors, the attempt probably increased the percentage of “no” votes in
other corridors because the vagueness of the overall plan was a central point
made by opponents.
In the Sonoma County case, the County Board of Supervisors added some
projects off the 101 corridor to try to increase the geographical equity. Since
the voting results off the 101 corridor were no different in 1998 than in 1990,
when all the funds went to the 101 corridor, this partial attempt at obtaining
geographical equity seemed to have little positive effect. It may have hurt by
increasing the cost of the transportation package by about 30% of the amount
to be financed by the sales tax.
Obtaining true geographical equity can be extremely difficult when multiple
corridors exist with significantly different degrees of congestion on each
corridor. In this situation, providing equal benefits in each geographic segment
based on tax revenues probably will be perceived as being inequitable by
people who use the more highly congested corridors. This situation existed in
both the Denver and Sonoma County cases. A viable solution in such a
situation might be to have at least part of the funding consist of user fees, such
as congestion pricing. Although this funding approach does not provide strict
geographical equity, it does result in having those who benefit from the
transportation improvements incur the costs of the improvements.
Finally, focusing on geographical equity over “true” transportation needs may
expose a proposal to the charge that it is based on “politics” rather than sound
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
115
planning principles. However, most respondents in Seattle and Santa Clara
were quite pleased with the soundness of the plans that were derived from an
equity approach. Whether addressing concerns about equity affects the
soundness of transportation proposals remains an open and an intriguing
question.
Determining the mix of highway and transit
Both Middleton and Beale et al. suggested that voter approval of transportation
projects with a transit component is increased by including a mix of highway
and transit, particularly when highway congestion exists. Transit only projects
may be perceived as offering limited congestion relief to the many voters who
would continue to drive, even with higher speed rail transit.
The four case studies examined do not support the contention that a reasonably
balanced mix of highway and transit improvements increases the probability of
voter approval. The transportation packages in two of the cases studied, Puget
Sound and Denver, consisted primarily of transit. One of these cases won by a
large margin and the other one lost by a large margin. In the two cases with a
somewhat balanced mix of transit and highway capacity increases, one won by
a small margin (Santa Clara County) and the second lost by a moderate margin
(Sonoma County). Of course, because these four cases were different in a
number of ways not pertaining to the mix of highway and transit, it is not
surprising that a strong correlation between the mix of highway to transit and
voting results has not been discovered.
As indicated in the previous paragraph, no systematic correlation exists in the
four cases between voter results and the ratio of highway to transit
improvements. However, a comparison of Puget Sound’s unsuccessful attempt
in 1995 to their successful attempt in 1996 suggests that a mix of highway to
transit can increase the probability of voter acceptance. The unsuccessful
attempt in 1995 was essentially a transit only package, while the successful
1996 package was modified to include a highway component. Yet, here also,
other factors changed from 1995 to 1996 in this case that could have caused the
different results.
Determining the specific mix of highway and transit improvements is made
more challenging by interest groups that have conflicting views on modes of
transportation, as well as different views on growth. Environmental groups
tend to be anti-highway and anti-growth while taxpayer groups tend to be antitransit. The difficulty of developing a transportation package satisfactory to
both anti-highway and anti-transit groups is demonstrated in the Sonoma
County case. Even though a citizen advisory committee was formed,
consisting of pragmatic representatives from both groups who made
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
116
Concluding Observations and Recommendations
compromises to satisfy the other group, opposition developed from both the
anti-highway and anti-transit groups.
The business community is another important interest group. This community
is the primary potential source of funds to run a campaign. It also can provide
important consulting support and leadership to campaign efforts. However,
this group does not make developing an acceptable mixed transportation
package more difficult. The business community in each case did not seem to
be either anti-transit or anti-highway. In addition, representatives in this group
seemed to advocate smart and controlled growth and also tended to have a
more pragmatic perspective than some representatives of environmental and
taxpayer groups.
Citizen advisory committees, incorporating both anti-transit and anti-highway
groups, were used in both Santa Clara County and Sonoma County. Yet, these
similar efforts generated different results in these two cases. The efforts were
successful in Santa Clara County but unsuccessful in Sonoma County. A
combination of luck and long-term coalition building seem to account for more
successful coalition building in Santa Clara County than in Sonoma County.
The EDF, which was a critical opposition group in the Sonoma County case,
did not decide until after the Santa Clara County election to get actively
involved in opposing transportation measures. Thus, EDF’s lack of
involvement in Santa Clara County can be attributed largely to happenstance.
At the same time, a longer-term strategy of coalition building existed in Santa
Clara County than in Sonoma County. In Santa Clara County, this attempt
started in the 1970s and was successful over time. On the other hand, coalition
building started in Sonoma County in 1990 and this attempt did not succeed.
In summary, an analysis of the impact of the highway to transit ratio on voting
results does not support any generalized optimal ratio. At the same time, the
ratio probably has a significant impact on voting results. Thus, careful case- or
situation-specific planning and research probably is needed to determine the
appropriate ratio of highway to transit improvements.
Funding sources
Each of the four cases relied on a combination of already existing state and
federal gas tax revenues and increases in the local sales tax of .004 or .005.7 In
each instance, voter approval was needed for the increase in the sales tax. Once
the decision is made to rely on an additional revenue source to the already
existing gas tax, a sales tax seems to be appropriate. The obvious advantage of
7
A motor vehicle excise tax rate increase was included along with a sales tax increase in the
Puget Sound case.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
117
a sales tax is a small cost per household, both because the tax is spread across
all households and because the aggregate expenditure base is quite large.
However, supplemental funding sources also should be seriously considered.
One additional funding source is considered here, namely the use of toll lanes
only on additional lanes constructed in the transportation package. This source
has been discussed and analyzed previously from a number of different
perspectives. The analysis perspective used here is the impact of toll lanes on
voter acceptance, both directly and also indirectly by influencing the nature
and extent of opposition to funding initiatives.
Using toll lanes as one version of user fees and congestion pricing was studied
in Sonoma County. As mentioned in the Sonoma County chapter, a study
commissioned by the Metropolitan Transportation Commission proposed this
as a serious option for Sonoma County. On the other hand, a survey conducted
in the county suggested that the toll lane concept would have a negative impact
on voter approval.
Given the limitations of survey methodology, it is premature to conclude on
the basis of survey results that a toll lane concept will have a negative impact
on voter acceptance of transportation funding initiatives. Although surveys
often can be used to measure attitudes and opinions, they may not provide
valid and reliable measures of attitudes that are not strongly held. Many
political issues have this characteristic, as evidenced by the rapid change in
some political attitudes discovered through tracking surveys. These attitude
changes occur because attitudes may not be strongly held and because voters
are often exposed to frequent arguments presented by proponents or
opponents.
If the negative attitudes of voters towards toll lanes, discovered in surveys,
may not be strongly held, these attitudes could improve over time if the issue is
framed properly. Negative attitudes may be based on perceptions that a toll
lane would result in long lines at a toll booth, that it would be an unavoidable
tax, and that it is unfair to tax drivers who depend on their cars and are already
paying high gas taxes. The first two of these perceptions could be modified
through reasonable counter-arguments. Technology exists to determine toll
charges electronically without toll booths. In addition, it would, in fact, be an
avoidable tax. Solo drivers would have the option of driving in already existing
congested lanes or driving in a newly constructed and less congested toll/HOV
lane.
Even if voter acceptance of transportation funding initiatives will not be
negatively influenced by the inclusion of a toll lane in the initiative, its
inclusion probably would not have a direct positive impact on voter
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
118
Concluding Observations and Recommendations
acceptance. However, it could have an indirect positive impact by minimizing
opposition from two groups that historically tend to oppose transportation
funding initiatives, namely taxpayer associations and environmental groups.
Interviews conducted with representatives from both groups in the Sonoma
County case, which is the only case in which organized opposition existed
from both groups, suggest that opposition would not have existed if the new
lanes on Highway 101 would have been a combination of HOV and toll lanes.
Of course, since these expressed views were based on a rather vague
hypothetical situation, it is difficult to say with any certainty whether either
group in the opposition coalition would have dropped out of the opposition if
the toll lane concept had been included.
To summarize the above discussion of using toll lanes, this concept could have
a neutral direct impact on voters but have a positive indirect impact by
minimizing opposition from environmental or taxpayer groups. Additional
research is needed to evaluate the wisdom of using the combination of toll/
HOV lanes as a supplemental funding source.
Marketing and communications approaches
Developing contingency plans for opposition or
extensive news media coverage
In the Puget Sound, Santa Clara County, and Sonoma County cases, a number
of apparently effective and appropriate steps were taken to both develop a
working coalition and to try to minimize effective opposition. At the same
time, when opposition did develop in the two unsuccessful cases, Denver and
Sonoma County, apparently no contingency plan existed for responding to
opposition. Since an important lesson of the cases is that opposition frequently
develops, developing a contingency plan to react to such opposition, as well as
possibly preempting this opposition, may be critical aspects of the campaign.
Contingency planning is needed when reaction times are short and such is the
case in political campaigns with transportation funding initiatives.
Even if opposition does not surface, the media, particularly newspapers, can
lessen voter support for a transportation package initiative, even with relatively
balanced coverage, depending on the issues of focus. Extensive newspaper
coverage existed in the Denver and Sonoma County cases, which almost
certainly lessened voter support. In both of these cases, the opposition focused
their efforts on obtaining extensive coverage since they had very little money
for advertising. Although their efforts probably were successful in influencing
the nature of the coverage, interviews with newspaper people suggest that
coverage would have been extensive in both cases even without the efforts of
organized opposition. Extensive newspaper coverage in Sonoma County
started months before the opposition surfaced. No information collected in any
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
119
of the cases suggests that the proponents had a contingency plan for extensive
newspaper coverage, whether the coverage was stimulated by opposition or
occurred without any competition. Efforts were made in Santa Clara County to
expose high level executives of the leading newspaper in the county to
successful light rail projects. A lesson learned, however, in both Sonoma
county and the Denver case is that the news division may provide extensive
balanced coverage even when leaders of the editorial group are strong
supporters of transportation measures.
Assuming that contingency plans should be developed to deal with both
effective opposition and extensive news coverage, the substance of the
contingency plan must be formulated. A starting point in developing a
contingency plan is anticipating the arguments that are likely to be used by
opponents or appear in the news. The cases in which opposition existed
suggest that many of these arguments can be predicted. Three such arguments,
1) high costs, 2) minimal impact on traffic congestion, and 3) double taxation,
are discussed below.
High costs
One opposition argument used in all three cases in which competition surfaced
was the high cost of the transportation package. This argument was frequently
reinforced in the newspapers, which focused on the total cost of the package
with very little mention of the cost per household. Obviously, proponents
should try to emphasize the cost per household rather than the total cost of the
package. It was reasonable for proponents to initially avoid any mention of
costs, since the packages were all expensive, with the least expensive one
being about $1 billion. Nevertheless, once opposition begins to emphasize
costs, proponents should seriously consider aggressive efforts to encourage
news sources to cite the cost per household rather than, or in addition to, the
total cost.
Limited highway congestion relief
Another common opposition argument of passenger rail is that it will not take a
sufficient number of drivers off the roads to have a significant impact on
highway congestion. A television advertising campaign in the Denver case
showed rush hour traffic on the most congested highway corridor both with
and without passenger rail on the corridor and the results were virtually
identical. It is not surprising that opponents focus on this issue. First, the
highway congestion relief issue is important to voters when tied to expensive
transportation packages with a significant passenger rail component, which
was a characteristic in all four cases studied. Second, since proponents focused
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
120
Concluding Observations and Recommendations
heavily on this congestion relief issue in each case, opponents basically were
forced to address this issue.
Given that voters are unlikely to approve an expensive transportation package
with a significant passenger rail component unless they perceive that highway
congestion will be lessened, it is critical to develop an appropriate response to
opponent claims of limited congestion impact or newspaper scrutiny of this
issue. Since proponent claims of congestion relief will usually be challenged
by opponents, proponents must have some logical basis and support for their
claims. The impact of transportation projects on traffic congestion usually is
projected in corridor studies commissioned by those developing the
transportation package. Thus, opponents can use these corridor studies to
identify unsubstantiated or vague claims. The Denver case provides an
example of this approach and it is likely to be used elsewhere by opponents
given its success in this case. Here opponents used corridor study predictions
of very limited travel time differences between the GTR package and the
baseline package to counter proponents’ claims that the passenger rail system
would provide traffic relief to automobile as well as to transit users. This case
also demonstrates that opponents are smart enough to realize that the critical
statistic for voters is reduction in travel time for drivers rather than proponents’
statistics such as the number of cars taken off the highways by the introduction
of a passenger rail system.
The impact of a transportation package on congestion relief depends in large
part on the nature of the package. Some mix of highway and transit
improvements, possibly combined with careful land use planning, is needed to
provide real relief in highway congestion. Survey research conducted in
Sonoma County after the election provides support for the integration of land
use planning and transportation planning. One of the arguments made by
opponents in this county was that an increase in highway expansion, without
corresponding growth restrictions, would result in population increases that
would negate the impact of increases in highway capacity.
Assuming, as discussed above, that proponents can support claims of
significant highway congestion relief resulting from the transportation
package, the response to this argument should depend upon the particular
combination of factors that exist. The plausible responses are identified below,
followed by a discussion of the different situations in which each response
probably should be used.
All plausible responses include making general claims of highway congestion
reduction in paid advertising. The responses differ in the identity of an added
element or elements. One plausible added element involves attacking
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
121
opponents for not having an alternative plan. This approach was used
successfully in the Puget Sound case. A second plausible element involves a
rather general response in the news media. Such a response would tend to
focus on successful uses of passenger rail in other counties and regions. This
approach was used unsuccessfully in the Denver case. As discussed below,
however, this approach could be successful in certain situations. A third
approach involves making more detailed responses and explanations in the
news media. These detailed responses probably would involve projections
from corridor studies, but there is a risk that the projections will be questioned
as happened in the Sonoma County case.
If opposition either does not have a specific plan or has an inferior plan, this
should be brought to the attention of the public. This approach can be
particularly effective when current highway congestion is perceived as being
severe, as was the situation in the Puget Sound case. In this situation, voters
probably tend to desire an immediate public solution, as long as the costs are
reasonable and the package will have an impact on congestion.
If the media does have even one article on the traffic congestion issue, some
response is needed and it should be prepared ahead of time since response time
will tend to be quite short. The most difficult decision is whether to provide a
general or specific response in the news media once it becomes evident that the
media will focus on this congestion relief issue. A closely related and difficult
decision is whether to proactively approach the news media and encourage the
media to focus on the traffic congestion issue. Some of the variables that
influence these two difficult decisions are the extent that voters seem to believe
this opponent’s claim of limited impact, the extent that corridor studies support
a case for significant reductions in travel time on the highway with the
proposed transportation package, the current amount of highway congestion,
voter perceptions of the cost of the transportation package, and the extent of
media coverage of the highway congestion issue.
A necessary condition for a detailed response is the quality of the information
existing to support the response, realizing that opponents and possibly the
media will challenge the information and predictions. If the predictions in
corridor studies are not inflated and if they incorporate acceptable modeling
techniques, data, and assumptions, then a detailed response might be
appropriate. Given, however, that any projections can be questioned, a detailed
response in the media should be used only when voters perceive that the
current package might be too costly, as was the case in both Denver and
Sonoma County, or when the existing level of highway congestion is not
sufficient in magnitude to cause voters to want an immediate solution.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
122
Concluding Observations and Recommendations
Proponents should approach the news media actively about the highway relief
congestion issue when voters
• have doubts about the transportation package’s impact on
congestion,
• do not perceive that current congestion is severe enough to create a
strong need for an immediate attempted solution, or
• are concerned about the package being too expensive.
When these conditions exist, voters may tend to opt for temporarily accepting a
less than severe status quo transportation system rather than taking a risk on an
unproven and costly transportation package. This voter tendency might be
magnified by a perception that rejecting a costly package with uncertain
benefits will tend to cause the county or region to resubmit a less costly
package. This situation seemed to occur in the Puget Sound case, in which the
1996 package was significantly less costly than the 1995 package which the
voters rejected. This situation also occurred in Denver, where the decision has
been made to build a light rail line, estimated to cost less than $1 billion, on the
most congested corridor only. This compares to the 1998 GTR package that
cost between $8 and $16 billion.
Countering charges of double taxation
Although the double taxation issue of a gas tax plus a sales tax was not an
explicit opposition argument in any of the three cases in which competition
surfaced, a post-election survey conducted in Sonoma County indicated that
this was an important issue to many “no” voters. Since anti-tax sentiments are
not as prevalent in Sonoma County as in many other counties and regions, this
double taxation concern may be quite common. As proponents in Sonoma
County realized after the election, this concern can be countered by indicating
that 1) the gas tax is not intended to fund all transportation projects, 2) gas tax
revenue is more likely to be provided to a county or region if a separate local
revenue source is used, and 3) when appropriate, what are the past and likely
future uses of gas tax revenues in the county or region. Trying to communicate
this message through the news media probably is more appropriate than doing
so through advertising for a couple of reasons. First, paid advertising probably
should focus on positives rather than countering negatives. Second, adequately
addressing the double taxation issue involves some complexity and detail. The
news media, particularly newspapers, provide a better means for presenting a
complex message than paid advertising.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
123
Dealing with the media
Extensive news media coverage existed in both of the unsuccessful cases. In
contrast, very limited news media coverage existed in both of the successful
cases. A simplistic conclusion from this pattern is that extensive news
coverage of transportation initiatives will result in a failed initiative while
limited coverage will result in a successful initiative. Although this conclusion
is overly simplistic, it may be appropriate for proponents to avoid taking steps
to encourage news coverage of transportation initiatives. Particularly when
proponents are able to raise significantly more campaign money than
opponents are, extensive news coverage will be the opponent’s most effective
alternative for presenting their message to the voting public. In addition,
proponents correctly realize that media news departments usually will attempt
to provide balanced coverage even when editorial departments strongly
support the transportation measures. Balanced coverage is likely to be viewed
negatively by proponents when they have a big spending edge. Proponents in
each of the four cases examined seemed to use this logic. No active attempts
were made in any of the cases to encourage coverage in the news media.
It may often be prudent for proponents to avoid encouraging news media
coverage. At the same time, as soon as it becomes apparent that coverage will
exist, they should consider 1) making strong efforts to influence the topics of
emphasis in the media, 2) having arguments ready to present to the media, and
3) having effective spokespeople who take the initiative in contacting the
media rather than waiting for the media to contact them. These recommended
steps were not taken in either of the two cases in which extensive coverage
existed. The failure to follow these steps is difficult to explain, given early
warning signs in both the Denver and Sonoma County cases that coverage
would be extensive even if opposition had not surfaced. Both newspapers in
Denver had news reporters whose primary assignment involved covering
transportation issues and coverage of the RTD Board was extensive prior to the
commencement of the campaign in the summer of 1998. Coverage in the
leading newspaper in Sonoma County also began months before opposition
surfaced.
A final issue that proponents could encourage the media to focus upon is the
magnitude of the problem with the existing transportation system combined
with projections of the problem exacerbating if no action is taken.
The degree of specificity of information presented by proponents
The transportation package in each of the four cases studied had multiple
components. This complexity requires a decision about the degree of
specificity to provide to voters concerning the nature of the package and cost
estimates.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
124
Concluding Observations and Recommendations
Both Beale et al (1996) and Middleton (1998) proposed that detailed
information be provided. The four cases studied in this research project suggest
that the relationship between the amount of specific information and voter
approval is more complex than described in earlier literature.
Specific information was not publicized in Santa Clara County and yet that
campaign was successful. In this case, proponents ran a “stealth” campaign.
Although avoiding specifics worked in this case, it is a risky approach that
depends on a combination of factors that often do not exist. Neither organized
opposition nor media scrutiny forced proponents to be more specific in this
case. In addition, since prior transportation packages, approved by voters, had
been completed in Santa Clara County, voters may have had some confidence
and trust in the planning process. When the stealth approach is used, it is
necessary to have a contingency plan to counter organized opposition and
media scrutiny.
The successful Puget Sound case demonstrates that determining the degree of
specific information is a complex decision. Both focus group and survey
research suggested that support for the funding measure decreases when
attention is focused on the details of the plan. Based on these results, the
majority of the proponents’ information campaign in this case involved
television advertising that did not present specific information. Proponents did
present specific information about the transportation projects and costs in an
eight-page direct mail piece. This was apparently done to demonstrate that a
detailed plan existed without calling attention to the details.
The Puget Sound case is important because it used a sophisticated and
integrated multi-media marketing communications approach, with different
messages and objectives in each medium. At the same time, this approach is
quite expensive and is dependent on the ability to raise a significant amount of
money. Since such funding may not exist, it is relevant to discuss a model
different than the Puget Sound one. One alternative model, not used in any of
the four cases, involves presenting detailed information on the Internet. The
interactive nature of this medium, along with increased public access, suggests
that it is an excellent means for presenting detailed and specific information on
project components and costs. Limited funds can then be used for media
advertising that does not focus on these details.
Miscellaneous issues
Mistakes were made in at least one of the unsuccessful cases that can be
identified and easily discussed due to the lack of complexity or controversy.
An underlying concept of the marketing and communications campaigns is the
need to do detailed planning, including contingency planning. Planning needs
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
125
to begin well in advance of the start of the campaign. Little evidence of
advanced planning exists in either of the unsuccessful cases. In each instance,
the transportation package was finalized only months prior to the election,
which provided little time for planning. In addition, to carefully control
indirect costs, political consultants were hired only a few months prior to the
election in both the Denver and Sonoma County cases.
If political consultants are hired more than a few months prior to the relevant
election, their fees will increase unless proponents share in the strategic
planning, ideally on a volunteer basis. Many people in the business community
have more planning expertise than political consultants. Identifying a team of
proponents a year or so prior to the election and charging the team with
developing a strategic campaign plan, including critical contingency plans, is a
reasonable step to consider. A team typically consists of local politicians with
expertise in the transportation area along with dedicated proponents from the
business community. Recruiting qualified volunteers should not be that
difficult due to a perception in the business community that a strong
transportation infrastructure is needed for economic development and
competitiveness.
Although strategic planning can be effectively conducted with a reasonably
large group, such is not the case with tactical planning. Even with
comprehensive contingency planning, some tactical adjustments will need to
be made during the course of the campaign. These adjustments usually need to
be made quite quickly. Very small groups, if not designated individuals, should
have responsibility for tactical planning during the campaign. Steering or
coordination committees used by proponents in both of the unsuccessful cases
were far too large to make effective tactical decisions quickly.
Opponents tend to prefer extensive media news coverage while proponents
prefer very limited coverage since the latter group almost always will have a
spending advantage. Although proponents cannot control the amount of
coverage, they may be able to reduce extensive coverage by presenting the
transportation package to the voters during an election in which many other
important issues are involved. Representatives from both major Denver
newspapers indicated that the amount of coverage of the Guide the Ride
measure would have been reduced if presented to the public during a
presidential election with a large number of competing issues and candidate
selections.
Some potential voter concerns are rather predictable, such as the extent that
passenger rail will reduce highway congestion, whether geographic equity
exists, and whether all projects will be completed because of costs being
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
126
Concluding Observations and Recommendations
underestimated or tax revenues being overestimated. Other voter concerns,
however, would have been difficult to predict during the campaign without
tracking research. One example of a more unpredictable issue was the doubletaxation issue that arose in Sonoma County. Another example from Sonoma
County was the issue of the potential impact of the transportation package on
population growth. It would have been easy to either overestimate or
underestimate the importance of this issue to voters without some sort of
tracking research. The basic point is that tracking research should be seriously
considered to help determine the need for tactical adjustments.
Summary and conclusions
Four county or regional cases have been examined. Each case involved a sales
tax of either .004 or .005 to fund a complex and expensive transportation
package that was presented to voters for approval or rejection. Voters approved
the package in two of the cases and rejected the package in the remaining two
cases. All of the cases were quite recent, with two occurring in 1996 and two in
1998. Each package had a significant passenger rail component and three of
the four had a significant highway component. In each case, proponents spent
at least $400,000 in a marketing campaign. Opposition also organized a
campaign in three of the cases, even though they had significantly less money
to spend than proponents.
In each case, attempts were made to identify characteristics of the coalition
building process, the transportation package and the process used to develop
the package, and the marketing and communications campaigns used. These
characteristics were studied to generate insight about factors influencing
voting results.
Some of the insights presented are consistent with those in the literature.
However, a number of insights not identified previously in the literature have
been generated by the current research. Many of these new insights involve the
marketing and communications campaign used by those trying to obtain voter
approval of a transportation funding measure. Insights about contingency
planning were developed in response to opposition or media scrutiny that had
not been presented in previous literature.
Some of the action and process implications of the case research have been
identified in this chapter. The action and process implications should be
interpreted with caution since a small number of case studies provide the
information base for these implications. The fact that two efforts failed in an
otherwise favorable environment, however, suggests that proponents of such
measures would do well to consider implementing these suggestions, because
any omission might be critical.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
127
Summary of action and process recommendations
• A budget cap should be identified to determine the magnitude of
the transportation package. This cap usually should be no greater
than projected sales tax revenues for a ten-year period at a sales tax
rate of no more than .005. Lower budget caps should be established,
depending on the magnitude of voters’ degree of dissatisfaction
with the existing transportation system, and voters’ attitudes
towards taxes. A budget cap usually will force developers of a
transportation package to prioritize potential components of the
package.
• Priorities should be based upon information from a variety of
sources. Corridor studies should be used but should not be the sole
source of information. In addition, information should be collected
directly from the public through research techniques such as focus
groups and surveys. Surveys can be useful in identifying voters’
priorities, as demonstrated in one of the successful cases. The
involvement of a citizen advisory group in the decision-making
process also can be useful in establishing priorities. Representatives
from key interest groups such as the business community and
environmental representatives, should be included in these citizen
advisory groups.
• In one successful case, priorities were established by identifying the
less critical segments of some passenger rail lines and eliminating
them from the transportation package. Additional cases need to be
examined to determine the feasibility of this incremental approach.
• A combination of highway and transit improvements should be
included in a transportation package whenever it is possible to do
so.
• Although opposition may not be visible when strategies are
planned, such opposition usually exists and it may come from
surprising sources. Since the campaigns usually are of short
duration, rapid responses to opposition are needed. Contingency
planning is needed to provide effective rapid responses. The most
common themes used by opposition are identified below along with
possible responses.
•
“The transportation package costs too much.” This can be
countered by framing the costs on a per household per year or
per month basis.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
128
Concluding Observations and Recommendations
•
“Gas taxes that voters are already paying should be used as the
primary funding source rather than creating an additional sales
tax.” This argument involves a more complex response. The
amount of gas tax revenue being used to fund the current
package should be identified along with suggesting that using
another funding source will tend to result in more gas tax
revenue being provided.
•
“Commute times on the highways will not be significantly
reduced by a passenger rail system because an insufficient
number of drivers will switch to passenger rail.” As discussed
in the body of this chapter, the response to this argument should
be different in different situations.
• The news media, particularly newspapers, should be used to
counter opposition. Advertising is not a good way of presenting
complex and detailed information, which may be needed in each of
the three situations identified. These efforts should include attempts
to have the media focus on issues favorable to proponents, such as
the amount of current congestion that exists with the current
transportation system.
• In some situations, the opposition should be criticized for not
presenting any solution, presenting only vague solutions, or
presenting inferior solutions. This approach is most effective when
voter dissatisfaction with the existing transportation system is very
high. In this situation, voters will be inclined to vote for an
immediate solution, even when they are uncertain about the extent
of improvement generated by the transportation package.
• Even when opposition does not exist, the news media, particularly
newspapers, may provide extensive coverage related to the
transportation package. Advanced planning needs to be conducted
for this contingency.
• Because of the complexity involved in developing appropriate
contingency plans, planning should begin about one year prior to
the election. Tracking research should be conducted when funding
permits, to help in both the strategic planning and the tactical
adjustments that often need to be made quickly during the course of
the campaign.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
129
Important and difficult issues involving the transportation package
Only one of the above enumerated items deals directly with desired
characteristics of the transportation package. Although it may not be reassuring
to those attempting to construct a package acceptable to voters, it was learned
that it is difficult to provide specific recommendations about characteristics of
an acceptable transportation package.
Some of these characteristics probably will not be resolved even if additional
research is conducted. Given the trade-offs with these characteristics, it is
unlikely that any reliable conclusions across situations can be determined. The
second recommendation above dealing with citizen advisory groups and
systematic voter research during the planning process is particularly critical,
given the difficulty of developing a general template for a transportation
package.
The geographical distribution of benefits
Prior research stressed the need to distribute transportation projects over
multiple corridors in an approximate proportion to the revenue raised along
each corridor. This recommendation has merit, both from the perspective of
voter and local political support. On the other hand, one of the lessons learned
in the four cases studied is that this recommendation is risky when the
transportation problems, such as congestion, are not equal across corridors.
The attempt in Denver to distribute benefits to each corridor demonstrates
these risks. First, this attempt resulted in a very expensive transportation
package. Second, this attempt was susceptible to opponent’s charges that some
of the corridor improvements were expensive and would have only limited
benefits.
The ratio of highway to transit improvements.
Although the cases suggest that voter approval is unlikely for an expensive
transportation package without increased highway capacity, no ideal ratio
seems to exist. The conflicting views of special interest groups towards
highway and transit complicate the process of determining a ratio.
Environmental groups tend to be opposed to highway capacity increases while
tax groups and some conservative groups, such as the Independence Institute in
the Denver case, usually oppose rail transit.
The nature of highway capacity improvements.
For a variety of reasons, including jurisdictional issues, highway capacity
increases in each of the cases studied involved HOV lanes. Voter attitudes
towards HOV lanes might be becoming more negative through experience. If
voters perceive that HOV lanes are underutilized and do not encourage many
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
130
Concluding Observations and Recommendations
people to switch to carpooling, which is not an unreasonable perception, voter
dissatisfaction with HOV lanes may increase. Three recent actions taken by
transportation agencies in California are consistent with this possible shift in
attitudes toward HOV lanes. First, Caltrans has requested a study of nonpricing methods to optimize HOV lane usage. These methods would allow
some solo drivers to use an HOV lane on a given day, with access determined
by such things as the last digit on a license plate. Second, the Metropolitan
Transportation Commission studied congestion pricing on HOV lanes in
Sonoma County. Third, the MTC also recently converted lightly used HOV
lanes on Interstate 580 in Richmond CA to unrestricted lanes (San Francisco
Chronicle, September 11, p.A20.).
Although highway capacity increases involved HOV lanes in the two
successful cases studied, Puget Sound and Santa Clara County, this approach
may be less successful in the future due to shifting public attitudes toward
HOV lanes. At the same time, any shift away from using highway capacity
increases for HOV use will likely provoke more organized opposition from
environmental groups. Thus, the extent that highway capacity increases
involving HOV is a difficult issue that will require careful analysis and
research in each individual situation.
Linking population growth restrictions with transportation funding
initiatives
Growth restrictions were not bundled with the transportation funding initiative
in any of the four cases studied. However, opponents argued in both of the
unsuccessful cases that the transportation improvements would do more to
stimulate population growth than to reduce congestion. Although post-election
surveys did not indicate that this was a primary concern for most opposing
voters, it was a concern expressed by some opposing voters in Sonoma
County. Particularly if a tax on development were used for partial funding of a
transportation package, including growth restrictions could reduce the tax
burden on current residents as well as generate support from no-growth and
slow-growth voters. On the other hand, this bundling of growth restrictions and
growth-related taxes with transportation funding initiatives could generate
opposition from growth-oriented segments of the business community.
Since conflicting views of growth exist, a decision to link growth restrictions
with transportation funding initiatives will require careful analysis and
research in each individual situation.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Concluding Observations and Recommendations
131
Use of an incremental approach
An incremental approach was taken in Santa Clara County to maintain
geographic balance while reducing the cost to a level that could result in voter
approval. This approach eliminated some segments of proposed rail transit
lines, with the expectation that the segments eliminated from the 1996 package
would be included in a future package. This approach has both positive
features and risks compared to the alternative of completing lines on the
highest priority corridors. It is an approach that should be analyzed and
researched in individual situations.
A final word
The cases studied have demonstrated that obtaining voter approval of
transportation packages is very difficult, particularly when a basic bus transit
system exists and the proposed transportation package is expensive. Even
when early polling identifies a significant margin of voter approval, the final
result may be rejection by a significant margin, as was the case in Denver.
Voting intentions can be quite unstable in this type of election, which is not
based heavily on liberal or conservative ideologies or partisan affiliations. The
case methodology used in this study demonstrates that a great deal can be
learned from this systematic research.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
132
Concluding Observations and Recommendations
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Abbreviations and Acronyms
ABBREVIATIONS AND ACRONYMS
Caltrans
California Department of Transportation
CEO
Chief Executive Officer
COST
Citizens Opposed to Sitting in Traffic
CTC
California Transportation Commission
CTR
Citizens for Traffic Relief
EDF
Environmental Defense Fund
EIR
Environmental Impact Report
GTR
Guide the Ride (Denver Campaign)
HOV
High Occupancy Vehicle
ISTEA
Intermodal Surface Transportation Efficiency Act
LOS
Level of Service
MIS
Management Information Systems
MPOs
Metropolitan Planning Organization
NWPRR
Northwestern Pacific Railroad
RTA
Regional Transit Authority (Central Puget Sound)
RTD
Regional Transportation District
STIP
State Transportation Improvement Program
SVMG
Silicon Valley Manufacturing Group
TABOR
Taxpayer’s Bill of Rights
TEA-21
Transportation Equity Act for the 21st Century
VMT
Vehicle Miles Traveled
VTA
Valley Transit Authority (Santa Clara Valley)
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
133
134
Abbreviations and Acronyms
BIBLIOGRAPHY
“A Tax Boost Even a Business Could Love,” San Jose Mercury News, 13
October 1996.
Advisory Commission on Intergovernmental Relations. Significant Features of
Fiscal Federalism: Volume 1: Budget Process and Tax Systems.
Washington: Advisory Commission on Intergovernmental Relations
M-197: 1995.
Altshuler, A. “The Decision-Making Environment of Urban Transportation,”
The Urban Transportation Problem. Cambridge: MIT Press, 1979.
Beale, H., Bishop, E. and Marley, W. “How to Pass Local Option Taxes to
Finance Transportation Projects,” Transportation Research Record 1558
(1996): 74-82.
Blumenauer, E. “ISTEA and Rail-Volution,” Intermodal
Transportation Efficiency Act 24, no. 3 (June 1998): 50.
Surface
“Both Sides Driven in Rail Campaign: Guide the Ride called a Boon,
Boondoggle,” Denver Post, 2 November 1997.
Brown, J. et al. “The Future of California Highway Finance.” MA thesis:
funded by the California Policy Seminar and the University of California
Energy Institute, Berkeley, 1998.
California State Board of Equalization. Tax Tips for Transit Taxes: Sales and
Use Taxes. Sacramento, CA.: 1985.
California State Controller's Office. “Government at a glance (1996-1997),”
www.sco.ca.gov/govglance/9697/gaagmap.htm Sacramento: California
State Controller's Office, (1999).
Calthorpe Associates Consulting Team. Sonoma/Marin Multi-Modal
Transportation and Land Use Study. Berkeley, CA: Calthorpe Associates,
1997.
“Campaign Near Road’s End: Foes Air Drive-Time Radio Spots,” Denver
Post, 31 October 1997.
Camph, D. “Transportation, the ISTEA, and American cities,”
www.transact.org/reports/mono/city.htm Washington, D.C.: Transportation Policy Project, n.d.
Carter & Burgess, Inc. Final Report: Southeast Corridor Major Investment
Study. Colorado: Carter & Burgess, Inc., 1997.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Bibliography
135
Central Puget Sound Regional Transit Authority. From No to Yes: Regional
High-capacity Transit for the Central Puget Sound Region. Central Puget
Sound: Central Puget Sound Regional Transit Authority, 1997.
Ciruli, F. The Loss of Transit in Metro Denver and the “Culture of
Opposition.” Denver: Ciruli Associates, 1998.
CIS Annual Legislative Histories. “Intermodal Surface Transportation
Efficiency Act of 1991,” Public Law 102, no. 240 (105 Stat. 1914): 291294.
Citrin, J. “How will it play in Peoria? (Or Monrovia? Or Milpitas?) What Do
Voters Want and What Are They Willing to Pay For?” Session 8,
Arrowhead Symposium, Los Angeles, California, January 1999.
Colman, S. “Letting the People Decide: Getting Voter Approval of
Transportation Sales Taxes.” Proceedings of the ITE District 6 Annual
Meeting, Reno, Nevada, 1987.
Commission on Transportation Investment. Final Report. California
Commission on Transportation Investment, January 1996.
Coogan, M. Transportation Politics in Atlanta: the Mass Transit Bond
Referendum of November 1968: with Comparisons to Referenda in Los
Angeles, San Francisco, Seattle, Washington, D.C. Cambridge, MA:
United States Department of Housing and Urban Development and
Harvard Law School. Urban Mass Transportation Study, 1970.
Demery, Leroy W., Jr. “A retrospect of rail financing votes in the U.S., 19621994: Part 1: Metropolitan regions outside of California.” Headlights vol.
58, no. 1-2 (1996): 3-12.
Federal Transit Administration. “Funding Strategies for Public Transportation.” Transit Cooperative Research Program Report 31, vol. 2.
Washington, D.C.: Federal Transit Administration, (1998).
Flem, L. and Schiermeyer, C. “Rail Transit: the People's Choice.” Railway Age
198, no. 9 (Sept. 1997): 71-80.
Forkenbrock, D. and Stoner, J. “Support for a Local Transit Tax.”
Transportation Research A vol. 18a, no. 3 (1983): 243-252.
Fulton, W. “Chapter 5, The Money Train,” The Reluctant Metropolis. Point
Arena, CA.: Solano Press Books, 1997.
“Gas Tax Forecast Triggers Debate,” Santa Rosa Press Democrat, 31 October
1998.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
136
Bibliography
Gerston, Larry N. “County Voters March to Own Beat,” San Jose Mercury
News, 10 November 1986.
Gogerty and Stark. Regional Transit Authority Campaign Summary, n.p.,
1997.
Gomez-Ibanez “Financing the Future: Setting an Agenda for the Next
Decade,” Session 9, Arrowhead Symposium, Los Angeles, California,
January 1999.
“Guide the Ride Derailed: Metro Voters Say No to Transit Proposal,” Denver
Post, 5 November 1997.
Harmatuck, D. “An Analysis of Bus Transportation Referenda Voting in
Madison, Wisconsin.” Ph.D. Dissertation, University of Wisconsin, 1973.
Hill, E. Developing and Funding an Efficient Transportation System. CA:
Legislative Analyst's Office, (5 March 1998).
Hoover, J. “How to Build Community Support for Transit.” Mass Transit 24,
no. 5. (Sept.-Oct. 1998).
“Hwy. 101 Toll Lanes Feasible, Study Concludes,” Santa Rosa Press
Democrat, 9 July 1998.
Jones, D. Urban Transit Policy: An Economic and Political History.
Englewood Cliffs, NJ: Prentice-Hall, 1985.
Jones, D. “Voting with Our Wheels,” Arrowhead Land Use Symposium, Los
Angeles, California., 1998.
Kemp, J. “Santa Barbara County Measure D: a California County Meets its
Transportation Funding Needs Through a Local Sales Tax,” Third National
Conference, Transportation Solutions for Small and Medium Sized
Communities: conference proceedings (1991): 253-267.
Legislative Analyst’s Office. “1996-97 Budget Analysis: Transportation,
Cross Cutting Issues.” www.lao.ca.gov/a96acc.html#A1 CA: Legislative
Analyst’s Office.(1996)
Levine, J., Park, S., and Wallace, R. “Public Choice in Transit Organization
and Finance: The Structure of Support.” Paper presented at the Association
of Collegiate Schools of Planning Conference, Pasadena, California,
November 1998.
“Light-Rail Ads Stress Commuter Appeal,” Denver Post, 16 October 1997.
“Measure B’s Road to Success,” San Jose Mercury News, 11 November 1996.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
Bibliography
137
Middleton, W. “Why Some Voters Say Yes; Why Some Say No.” Railway Age
199, no. 2. (February 1998): G1-G8.
Pagano, Michael A., Devolution of Federal Aid Highway Programs: Cases in
State-Local Relations and Issues in State Law. Washington, D.C.:
Advisory Commission on Intergovernmental Relations, 1988.
Smerk, G. The Federal Role in Urban Mass Transportation. Bloomington:
Indiana University Press, 1991.
Sprague & Lewis. Federal Transportation Policy and the Role of Metropolitan
Planning Organizations in California. San Francisco, CA: Public Policy
Institute of California, April 1997.
“Sprawling, Sprawling,” Newsweek, 19 July, 1999, p. 25-29.
Taylor, B. When Finance Leads Planning: the Influence of Public Finance on
Transportation Planning and Policy in California. Ph.D. Dissertation,
University of California, Berkeley, 1992.
Taylor, Brian. "Public Perceptions, Fiscal Realities, and Freeway Planning:
The California Case." Journal of the American Planning Association 61,
no. 1 (1995): 43-56.
Taylor, Brian D. and William S. McCullough. “Lost Riders.” Access 13
(1998): 26-31.
“Traffic: Slow and Slower,” Seattle Times, 1 November 1996.
Trent, R. B. and Bernard, H. R.. Journal of Advanced Transportation 19, no. 3
(1985): 237-249.
U.S. DOT. “Public Involvement Techniques for Transportation Decisionmaking.” U.S. DOT (September 1996). www.fhwa.dot.gov/reports/pittd/
cover.html.
University of California, Berkeley. Institute of Governmental Studies.
[California general election materials, 1986 (Nov. 4), Alameda county
Measure B and Contra Costa County Measure C, sales tax increase for
transit projects; newspaper clippings and ephemera], 1986.
“Valley Firms Heavily Fund Effort to Pass Transit Tax,” San Jose Mercury
News, 30 October 1996.
Wachs, M. The Evolution of Transportation Policy in Los Angeles. n.p., 1977.
Weiner, E. Urban Transportation Planning in the United States, an Historical
Overview. Washington, D.C.: U.S. Department of Transportation, 1986.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
138
Bibliography
Womack, James and Pucher, John. The Urban Transportation System: Politics
and Policy Innovations. Cambridge, MA: MIT Press, n.d.
Zoller, T and Capizzano, J. Evolution and Devolution: a National Perspective
on the Changing Role of Metropolitan Planning Organizations in
Areawide Intermodal Planning. Virginia Department of Transportation,
June 1997.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
About the Authors
139
ABOUT THE AUTHORS
PETER HAAS, PH.D., PRINCIPAL INVESTIGATOR
Peter J. Haas is a Professor in the Department of Political Science at San José
State University. The author or numerous scholarly and professional articles,
he recently co-authored a text on policy research and was team leader for the
1998 Mineta Transportation Institute publication, Capital versus Operating
Grants for Transit: Economic Impacts for California.
KRISTEN SULLIVAN MASSEY, M.P.P.
Kristin Massey recently earned a Masters degree in Public Policy with a
concentration in transportation from the University of California, Los Angeles.
She also has a Masters Degree in Applied Cognitive Psychology with an
emphasis in driver behavior from the Claremont Graduate University. After
her participation in the current research was complete, she accepted a position
with the Office of Inspector General in the U.S. Department of Transportation.
LINDA O. VALENTY, PH.D.
Linda O. Valenty is Assistant Professor of Political Science at San José State
University. Dr. Valenty’s research has focused upon public policy (including
environmental, telecommunications, and transportation policy), psychological
profiling, behavioral prediciton, and conflict resolution.
RICHARD WERBEL, PH.D.
Richard Werbel is a Professor of marketing at San José State University. Prior
to his current focus on transportation, he conducted research in a number of
application areas, generally research methodology. He has published numerous
articles and papers including some that have received outstanding paper
awards.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies
140
Pre-Publication Peer Review
PRE-PUBLICATION PEER REVIEW
San José State University, of the California State University system, and the
Mineta Transportation Institute Board of Trustees have agreed upon a peer
review process required for all research published by the Institute. The purpose
of the review process is to ensure that the results presented are based upon a
professionally acceptable research protocol.
Research projects begin with the approval of a scope of work by the
sponsoring entities, with in-process reviews by the Mineta Transportation
Institute Research Director and the project sponsor. Periodic progress reports
are provided to the Research Director and the Research Associate Policy
Oversight Committee (RAPOC). Review of the draft research product is
conducted by the Research Committee of the Board of Trustees, and may
include invited critiques from other professionals in the subject field. The
review is based on the professional propriety of the research methodology.
Norman Y. Mineta International Institute for Surface Transportation Policy Studies